Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Migration to less popular rural areas
The characteristics, motivations and search
process of migrants
This research was funded by the Mansholt Chair for Rural Development (University
of Groningen).
ISBN 978‐90‐367‐5992‐2
ISBN (E‐publication) 978‐90‐367‐5993‐9
English correction: Language Centre of the RUG
Cover design: Helga de Graaf, info@eye‐candy.nl, www.eye‐candy.nl
Cover photo: Harry Cock, www.harrycock.nl
Printed by: Ipskamp Drukkers BV, Enschede, the Netherlands
© Rixt A. Bijker, 2013
All rights reserved. Save exceptions by the law, no part of this publication may be
reproduced in any form, by print, photocopying, or otherwise, without the prior
written permission from the author.
RIJKSUNIVERSITEIT GRONINGEN
Migration to less popular rural areas The characteristics, motivations and search process of migrants
Proefschrift
ter verkrijging van het doctoraat in de
Ruimtelijke Wetenschappen
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. E. Sterken,
in het openbaar te verdedigen op
donderdag 17 januari 2013
om 14.30 uur
door
Rixt Anke Bijker
geboren op 26 juni 1982
te Leeuwarden
Promotor: Prof. dr. D. Strijker
Copromotor: Dr. T. Haartsen
Beoordelingscommissie: Prof. dr. M. van Ham
Prof. dr. P. McCann
Prof. dr. C.H. Mulder
Voorwoord
Dit proefschrift betekent de afsluiting van een intensieve, leerzame en leuke periode
van onderzoek doen. Hoewel ik erg blij ben nu dit voorwoord te kunnen schrijven,
denk ik dat ook voor een promotieonderzoek geldt dat de weg naar het einddoel
belangrijker is dan het doel op zich. Die weg was niet altijd gemakkelijk, maar ik ben
blij dat ik hem heb afgelegd. Er zijn verschillende mensen die het afleggen van die
weg een stuk makkelijker en in ieder geval ook veel leuker hebben gemaakt.
In de eerste plaats zijn dat mijn promotor en copromotor, die, de een wonend
in een minder populair en de ander in een populair plattelandsgebied, zorgden voor
een goed evenwicht in de begeleiding. Dirk, heel erg bedankt voor het vertrouwen
en de ruimte die je me gegeven hebt om het onderzoek te doen. Tijdens de periode
van dataverzameling heb ik je praktische en pragmatische inbreng erg gewaardeerd.
Je goede en gedetailleerde feedback maakte het tijdens het schrijven mogelijk
stukken echt te verbeteren, dat heb ik als erg nuttig en leerzaam ervaren. Het was
erg fijn dat je altijd bereid was tijd vrij te maken om mee te denken als ik daar
behoefte aan had. Tialda, bedankt voor onze fijne samenwerking. Je inhoudelijke
expertise op het gebied van platteland, representaties en migratie heeft veel
bijgedragen aan het onderzoek en met je belangstelling en enthousiasme wist je me
altijd weer te motiveren. Ik bewaar goede herinneringen aan onze gezamenlijke
congresbezoeken aan Praag en Londen.
Het onderzoek was niet mogelijk geweest zonder de medewerking van de
zeven gemeenten waar ik een enquête onder nieuwe inwoners heb gehouden: De
Marne, Menterwolde, Reiderland (nu Oldambt), Ferwerderadiel, Wymbritseradiel
(nu Súdwest–Fryslân), Aa en Hunze en Westerveld, erg veel dank daarvoor. Ook
wil ik de respondenten bedanken die ik heb mogen interviewen over hun zoektocht
naar een nieuwe woning en die daarna bereid waren soms gedurende lange tijd
regelmatig een vragenlijst over de voortgang van hun zoekproces in te vullen.
Verschillende makelaars hebben ervoor gezorgd dat ik in contact kon komen met
deze huizenzoekers, ook aan hen ben ik dank verschuldigd.
Ik heb mijn tijd bij de Faculteit Ruimtelijke Wetenschappen altijd als erg leuk
ervaren. De basiseenheid Culturele Geografie vormde daarbinnen een fijne
uitvalsbasis. Een aantal mensen wil ik in het bijzonder noemen. Mirjam, vanaf het
begin van ons promotietraject hebben we een kamer gedeeld, misschien is het wel
symbolisch dat we nu heel even weer een beetje kamergenoten zijn. Maar behalve
een kamer hebben we veel meer gedeeld in veel goede gesprekken, zowel op het
werk als daarbuiten. Bedankt voor je luisterend oor en goede raad, en voor je vaak
terugkerende motiverende vraag in de ochtend: ‘weet je al wat je wilt gaan doen
vandaag?’. Ik ben erg blij dat je m’n paranimf wilt zijn! Heike, ook met jou als
paranimf ben ik erg blij, met zo’n stipt en organisatorisch sterk duo moet het goed
komen! Ook bij jou kan ik altijd terecht om dilemma’s of tegenvallers te delen, maar
gelukkig ook voor gesprekken over een breed scala aan niet‐werk gerelateerde
onderwerpen. Je vrolijkheid en enthousiasme werken aanstekelijk en met jou is het
altijd gezellig, of het nu op een terras aan de Zonnelaan is of met een biertje in de
Spieghel.
Petra, het is fijn om met jou het hele aio‐traject gedeeld te hebben. Ik herinner
me nog onze gesprekken samen met Mirjam in de stationsrestauratie in Utrecht vol
verwarring over waar het naartoe moest met onze onderzoeken. Gelukkig is het met
ons beiden helemaal goed gekomen! Naast onze interesse in onderzoek delen we
ook een interesse in muziek en dan nog wel in het bijzonder die van Bruce
Springsteen. Ik bewaar goede herinneringen aan ons Pinkpopavontuur in een tent
zonder haringen en ook aan geslaagde Eurosonicavonden. Marianna, my Greek
party girl, and of course rural girl! Ik mis je enthousiasme en hartelijkheid nu je in
Schotland woont. Dankjewel voor je steun en gezelligheid op en buiten het werk. Ik
denk nog met plezier terug aan ons gezamenlijke debuut op een wetenschappelijk
congres in Plymouth. Viktor en Sierdjan, bedankt dat ik jullie altijd kon lastig vallen
met een vraag of twijfel over een steekproef of een regressie analyse. Ook heb ik met
jullie en met onder meer Sander, Nora, Inge, Elen, Niels, Candice, Karen, Chris,
Gerd, Chris, Marije, Koen, Billie, Debbie, Linden, Sanne en Taede een hele gezellige
tijd gehad, met heel wat leuke lunches en vrijdagmiddagborrels die ik niet had
willen missen.
Buiten de faculteit wil ik Anna, Elvira en Marieke noemen, ik ben blij met jullie
als vriendinnen! Anna, dankewol foar us goede (en lange!) gesprekken, dyn
belutsenheid en enthousiasme en in protte geselliche jûnen en net te ferjitten tige
slagge fakansjes. Elvira, je hebt de afgelopen jaren veel moeten aanhoren over de
ups maar zeker ook de downs van het doen van een promotie‐onderzoek. Dit
gebeurde tijdens relaxte avondjes op de bank, al dan niet kijkend naar Boer Zoekt
Vrouw, maar ook tijdens fijne fiets‐ en schaatstochten. Het is erg fijn om te weten dat
ik altijd op je kan rekenen. Marieke, we hebben veel gezellige avonden thuis of in de
stad doorgebracht. Ik kijk ook met veel plezier terug op onze vakanties samen én
veel mooie schaatsavonturen. Bedankt voor het aanhoren van mijn verhalen, jouw
nuchtere blik werkt altijd lekker relativerend.
Sonja, Jeltje, Marijke, Nynke, Tjitske, Margriet, Cristina, Tjitske, Rixt, Afke,
Martha, Liesbeth‐Marije, bedankt foar hiel wat geselliche jûnen mei goede
gesprekken of soms just hearlik prate oer neat en domme grapkes meitsjen. Myn
reizen mei jim ha my brocht nei sear utienrinnende bestimmingen, fan Volendam
oant Uganda, mar altyd koenen we sizze: wat binne we moai fuort!
Marieke, we hebben elkaar de afgelopen tijd minder gezien dan ik misschien
zou willen, mede door mijn proefschriftdrukte, maar ik vind het erg fijn allerlei
promotieperikelen en veel andere dingen met je te kunnen delen. Ook de andere
socio’s, Jaantje, Marloes, Andrea en Martin, wil ik hier noemen, erg leuk dat we via
onze socio‐dates elkaar nog steeds zien.
De maandagavond is gereserveerd voor Harmonie’67 en dat betekent een
avond mooie muziek maken met z’n allen, maar ook veel gezelligheid en het
ontmoeten van leuke mensen. En als het programma van de dag erna het toelaat nog
een drankje in de Klikspaan…
Mijn ouders wil ik bedanken voor hun grote steun en betrokkenheid. Heit en
mem, jim wisten my altyd te motivearen om wer oan de slach te gean as it even net
woe. Mar jim ha ek altyd beneidrukt dat der mear is dan allinnich it skriuwen fan in
proefskrift. It is hiel fijn om te witten dat jim der altyd foar my binne. Dankewol!
Rixt Bijker ‐ Groningen, november 2012
Table of contents
1. Introduction 11
2. More than counter‐urbanisation: Migration to popular and
less‐popular rural areas in the Netherlands 25
3. Migration to less‐popular rural areas in the Netherlands:
Exploring the motivations 55
4. Different areas, different people? Migration to popular and
less‐popular rural areas in the Netherlands 85
5. Using a diary approach to explore the residential search
process of movers to rural areas 111
6. Conclusions 147
Appendix A: Survey 167
Appendix B: In‐depth interview guide 175
Appendix C: Electronic questionnaire 179
Summary in Dutch 189
Four chapters included in this PhD dissertation are reprinted from the following
publications and manuscripts:
Chapter 2
Bijker RA, Haartsen T. 2012. More than counter‐urbanisation: Migration to popular
and less‐popular rural areas in the Netherlands. Population, Space and Place 18: 643‐
657.
Chapter 3
Bijker RA, Haartsen T, Strijker D. 2012. Migration to less‐popular rural areas in the
Netherlands: Exploring the motivations. Journal of Rural Studies 28: 490‐498.
Chapter 4 (with some minor adjustments)
Bijker RA, Haartsen T, Strijker D. Different areas, different people? Migration to
popular and less‐popular rural areas in the Netherlands. Population, Space and Place.
DOI: 10.1002/psp.1741. (available online in ‘early view’).
Chapter 5
Bijker RA, Haartsen T, Strijker D. Using a diary approach to explore the residential
search process of movers to rural areas.
Submitted to an international journal.
11
1. Introduction
1.1 Motivation
‘The Dutch romanticise the countryside, with pastures full of grazing cattle: every
year 160,000 urbanites move to quietness and space’1, claims a newspaper headline
of an article summarising the findings of a report about the different uses and
representations of the Dutch countryside (see Steenbekkers et al., 2008). This article
and the report itself illustrate the transformation of the rural from a production to a
consumption space by describing the idyllic image held by respondents and
discussing developments connected to moving to or living in rural areas, recreation
and opportunities for small businesses (see also Markantoni, 2012; Woods, 2005). It
appears that one in three of the respondents in the report dreamt of moving to the
rural at some point. Migration into rural areas is often referred to as
counterurbanisation. Attention to this phenomenon started in the 1970s, leading to a
large amount of research (e.g. Beale, 1975; Berry, 1976; Champion, 1989; Boyle and
Halfacree, 1998), which eventually resulted in the dominant understanding of
counterurbanisation as the movement of middle‐class groups from the city in search
of new lives in an idyllic rural setting (Halfacree, 2008). After thirty years, research
into counterurbanisation seemed to have reached some kind of saturation point. It
was felt that everything that could be said had already been said about this
phenomenon (Ibid.). However, recently a ‘critical re‐appraisal’ of the concept of
counterurbanisation has started (see e.g. Grimsrud, 2011; Halfacree, 2001, 2008, 2012;
Milbourne, 2007; Smith, 2007; Woods, 2011). This thesis aims to contribute to this
critical reconsideration of migration into rural areas.
The reconsideration in literature of the dominant understanding of
counterurbanisation is based on empirical results as well as on more theoretical
papers. First, it is increasingly being acknowledged that rural areas differ in their
1 Nederlanders romantiseren het platteland, vol koeien in de wei, Trouw 21‐10‐2008.
12
popularity as places to live. The presence of amenities and accessibility are
important factors influencing this popularity (Argent et al., 2007; McGranahan, 2008;
Woods, 2005). Further, it is recognised that in addition to the pull of the rural idyll,
other motivations play a role, independent of the rural setting, such as the proximity
of family and friends or work‐related reasons (e.g. Grimsrud, 2011). In addition,
other groups move to rural areas along with the often mentioned middle class, such
as low‐income groups and migrant workers (Foulkes and Newbold, 2008; Hoggart,
2007; Woods, 2011). Milbourne (2007) points to the importance of moves within
rural areas over short distances. Halfacree (2012) draws attention to more temporal
mobilities in the rural environment. Finally, it appears that the traditional
conceptualisation of counterurbanisation, developed in more densely populated
countries, applies less to more sparsely populated countries (Grimsrud, 2011).
However, while it is increasingly being acknowledged that rural areas differ in
their popularity as places to live in, this notion has not yet translated to the research
studying the characteristics and motivations of rural in‐migrants. Research into the
characteristics and motivations of in‐migrants often focuses on those rural areas that
are attractive to live in, while migration into less‐popular rural areas remains
relatively under‐researched. This is also true in the Netherlands. Research has
investigated the characteristics and preferences of people considering moving to
rural areas in general (Heins, 2002, Van Dam et al., 2002) and the features of in‐
migration flows to rural areas in the Netherlands as a whole (Steenbekkers et al.,
2008) and specifically to amenity‐rich rural areas (Elbersen, 2001).
By studying in‐migration to less‐popular rural areas in the northern
Netherlands, this thesis wants to add to the development of an empirically based
broader understanding of counterurbanisation. This is not just relevant to the
scientific debate but it can also provide useful input for policymakers. Like other
rural areas in Europe, population decline has begun in the Netherlands, specifically
in peripheral parts of the country including the northern Netherlands (Haartsen and
Venhorst, 2010). In‐migration is one of the factors influencing population decline in
addition to births, deaths and out‐migration. Therefore, it is important to know
more about the characteristics and motivations of the people who move to these
13
kinds of areas and their search strategies, and especially about the different
migration flows to different types of rural areas (Haartsen, 2012). This information is
crucial when developing housing policies. It is important for policymakers in rural
areas to acknowledge that ‘the’ rural migrant does not exist. It is vital to recognise
that there are different groups of migrants with different motivations, also to create
more realistic expectations of the opportunities and benefits that in‐migration brings
to rural areas.
In the following section, a short overview of the literature discussing the
central concepts in this thesis is provided, followed by the research aim, the research
questions and an outline of the thesis.
1.2 Migration to rural areas
1.2.1 Popularity of rural areas
The finding that rural areas differ in their popularity as places to live in is referred to
by Woods (2005) as the ‘regionally uneven nature of counterurbanisation’. The
popularity of rural areas for migration purposes is found to be mainly influenced by
their accessibility and the presence of amenities (Argent et al., 2007; Bollman and
Briggs, 1992; Johnson and Beale, 1994; McGranahan, 2008). In addition, perceived
lifestyle and employment opportunities play a role (Johnson and Beale, 1994).
In the context of accessibility, transport availability plays a key role. For
instance, impressive rural repopulation has been found in accessible rural areas in
Ireland, along the road and railways, while the population is declining in less
accessible rural areas, particularly in the west (Gkartzios and Scott, 2009). Proximity
to cities also plays a role in the accessibility of rural areas. In the United States ,
Canada and Sweden, population growth is concentrated in the rural districts closest
to metropolitan and regional urban centres, while the more remote rural regions
suffer stagnation or decline (Bollman and Briggs, 1992; Johnson and Beale, 1994;
Westlund, 2002).
Amenities in a rural context most often refer to attractive landscape features or
climate. Regarding landscape features, this can be very specific landscape elements,
14
such as the coast (Argent et al., 2007), mountains (Shumway and Otterstrom, 2001)
or national parks (Elbersen, 2001). However, landscape has also been found more
generally to influence migration. People are most drawn to areas with a mix of forest
and open land, bodies of water, topographical variety and relatively little cropland
(McGranahan, 2008). These results correspond with the more general findings on
landscape preferences that people strongly prefer park or savannah‐like landscapes
with scattered trees or small stands, open vistas, even and uniform ground texture,
and a water source (Ulrich, 1986). These preferences appear to be largely
independent of culture, with the exception of people dependent on other landscapes
for their livelihood, such as farmers and foresters (Van den Berg et al., 1998; Yu,
1995). Preferences for semi‐open landscapes can be explained by the so‐called
prospect‐refuge theory, which argues that, the ability to see (prospect) without being
seen (refuge) increases perceived safety, which in turn increases the aesthetic
pleasure experienced (Appleton, 1975).
1.2.2 Characteristics and motivations of in‐migrants to rural areas
As mentioned earlier, rural in‐migration in general is often associated with the
movement of middle‐class families from the city, who are either retired or commute
to nearby urban centres for employment (Smith and Phillips, 2001; Stockdale, 2006).
When focusing on the limited evidence on migration into less‐popular rural areas, a
picture emerges that is different from this ‘classic’ image. With regard to the origin
and age of the movers, Stockdale (2006) found that most in‐migrants in
depopulating, peripheral areas in rural Scotland moved short distances and that
there was a mix of moves from cities, from smaller settlements and between similar‐
sized rural settlements. Most respondents in her study had moved before turning
fifty. Furthermore, less‐popular areas do not seem to attract the middle class. In
Stockdale’s study, skilled manual, trade‐type occupations dominated, with few
being employed in the professional or managerial sectors. Furthermore,
impoverished rural areas in the US, characterised by ample availability of cheap
rental housing, attract large groups of migrants with low incomes, varying from
working poor (Foulkes and Newbold, 2008) to welfare benefits recipients (Fitchen,
15
1995). Nonetheless, a more diverse group of in‐migrants were found moving into
fringe areas in Denmark, i.e. areas experiencing population decline and a low level
of economic activity (Andersen, 2011). In addition to some groups with quite low
incomes that Andersen assumes migrate for lower housing costs, these areas also
attract movers with middle or high incomes, partly consisting of couples with
children moving to detached houses, something they are probably unable to afford
in Denmark’s growth areas.
An idea has developed in recent years in housing research that traditional
background characteristics (i.e. income, education or age) may no longer be
sufficient to explain residential preferences, owing to a trend towards greater
differentiation in housing behaviour (see e.g. Heijs et al., 2009, 2011; Jansen, 2011).
This has resulted in approaches which explore the motives underlying consumer
preferences, in order to improve the understanding of housing preferences and
preferences for residential environments (Bettman, 1979; Jansen, 2011; Kersloot and
Kauko, 2004; Rokeach, 1973). However, in migration research little attention has
been paid to underlying motivational factors such as values. In Chapter 4 of this
dissertation we include value orientations in our analysis to characterise movers to
different types of rural areas.
The motives for migration to rural areas in general can be subdivided into
three groups: residential, household, and work and education motives (Van Dam et
al., 2002). The first group, residential motives, refers to the quality of the housing
and residential environment. The quality of the residential environment is often
regarded as rural areas’ most important pull factor (Deller et al., 2001; Gkartzios and
Scott, 2009; Halfacree, 1994; Halliday and Coombes, 1995; Walmsley et al., 1998). In
addition to the residential environment, housing characteristics can also be an
important reason for moving to the countryside, for example for a better or bigger
house, or a large garden (Gkartzios and Scott, 2009). The second group of motives,
the household motives, refers to moves motivated by life events, such as moving in
together, getting married, having children or getting divorced (Van Dam, 2000). The
desire to live closer to family also belongs to this category of motives for moving.
The third group of motives is related to employment and education. Migration
16
motivated by a change of job, the desire to live closer to work or to get an education,
fits into this category (Van Dam, 2000). A special motive for rural migration is return
migration (see e.g. Bolton and Chalkley, 1990; Elbersen, 2001). In addition to return
migration to a specific place, more general ‘return to the rural’ types of in‐migration
have also been found (Feijten et al., 2008; Gkartzios and Scott, 2009).
Looking at the motivations of movers to less‐popular areas, it seems that
instead of the quality of the residential environment, other motivations for choosing
the destination area are more central in the decision process. Studies have shown the
importance of more personal motivations, such as marriage and the proximity of
family and friends, employment considerations and housing reasons, in some cases
in particular related to the availability of affordable housing (Fitchen, 1995; Foulkes
and Newbold, 2008; Grimsrud, 2011; Stockdale, 2006).
However, existing studies have yielded inconsistent results with respect to the
importance of the pull of the countryside as a way of life for in‐migration to less‐
popular rural areas. While Stockdale (2006) found quality‐of‐life considerations to be
of little importance, the in‐depth interviews conducted by Foulkes and Newbold
(2008) reveal that in addition to the attraction of low housing costs, the rural
character of a community did play a role in the decision to move there. Moving to
these communities represented an opportunity to satisfy the desire to live in a rural
setting (Foulkes and Newbold, 2008). The extent to which the typical ‘quality‐of‐life’
motivations connected to the counterurbanisation concept also hold true for
migrants to less‐popular rural areas is thus not clear.
1.3 Research aim, research questions and thesis outline
The objective of this dissertation is to gain more insight into migration to less‐
popular areas. In particular, the focus is on the characteristics and motivations of the
in‐migrants. To determine to what extent the outcomes are specific for less‐popular
areas, a direct comparison with in‐migrants to popular areas in the same region is
also included in the study. The popularity of rural areas as places to live is defined
in this study using the average house price per municipality. This seems to be the
17
most appropriate indicator in the Dutch context (see also Chapter 2). Within the
Netherlands, the North was chosen as a study area, as it can be considered the most
rural part of the Netherlands based on address density and the perceptions of the
Dutch people (Haartsen et al., 2003).
Three research questions were formulated based on the research objective:
1. What are the characteristics and motivations of movers to less‐popular rural
areas?
2. To what extent do the characteristics and motivations of movers to less‐
popular
and popular rural areas differ?
3. How do potential movers search for a place to live in rural areas?
These questions are addressed in Chapters 2, 3, 4 and 5. In Chapter 2 three
types of rural area are distinguished within the northern Netherlands, based on
differences in their popularity as places to live in. The popular, less‐popular and
average rural areas, which are defined based on average house price per
municipality, are characterised using the characteristics that in an international
context are known to influence the popularity for living of an area: accessibility and
the landscape. After that, their migration patterns are compared using data from
Statistics Netherlands. Some general migration statistics for the study region are
provided. Finally, secondary data from Housing Research of the Netherlands are
used to make a first exploratory comparison of the characteristics and motivations of
migrants to popular and less‐popular areas using a multinomial logistic regression
model.
The exploratory comparison in Chapter 2 yielded some differences between
movers to less‐popular and popular areas. However, the available secondary data
used in the analysis appeared to be insufficient to enable an effective comparison.
For instance, information on return migration was lacking and motives for moving
were mixed with motives for location choice. Therefore, new survey data were
collected for the analysis in Chapter 3. Of the less‐popular municipalities defined in
18
Chapter 2, four were selected as study areas. Using the data (N = 664) we collected in
these municipalities, we investigate the characteristics and motivations of the in‐
migrants, which answers the first research question. We shed more light on the
motivations for moving to less‐popular areas by disentangling them into reasons for
leaving the previous place of residence, motivations for moving to the rural in
general and motivations for moving to the specific less‐popular rural area in
question (cf. Bolton and Chalkley, 1990). In addition, we take the
multidimensionality of motives for moving into account, which makes it easier to
reveal the importance of the rural character of the destination (Halfacree, 1994;
Halliday and Coombes, 1995). We use logistic regression analysis to combine the
movers’ characteristics with their motivations. In doing so, we explore which
motivations are mentioned by whom.
After specifically focusing on less‐popular rural areas, a direct comparison is
made in Chapter 4 between movers to less‐popular rural areas and to popular rural
areas. To do this, survey data were collected in three municipalities characterised as
popular in Chapter 2 (total N = 1717) in a similar way to the less‐popular areas in
Chapter 3. Compared to the analysis in Chapter 2 using secondary data, the
motivations for location choice are now separated from the motives for leaving the
previous residence. In this analysis we focus specifically on the motivations for
choosing the destination area. We also include return migration and incorporate a
more detailed description of the movers’ previous residence. In addition, as well as
characterising the movers using ‘traditional’ sociodemographic characteristics, we
also use their values. These underlying motivational factors have been used in
housing research to improve the understanding of housing preferences (e.g. Coolen
and Hoekstra, 2001; Coolen et al., 2002). Values have received little attention so far
in migration research. We explore whether they can also aid in distinguishing the
movers to different types of rural area. We include the values along with
background characteristics and motivations in a logistic regression analysis, which
allows us to investigate which of these variables predict a move to a specific type of
rural area, while controlling for the effect of other factors.
19
While Chapters 2, 3 and 4 focus on the migrants who already made a move to
the rural, Chapter 5 concentrates on the process by which potential movers look for
places to live in rural areas. Little attention has been paid so far to the residential
search process in a rural context. This study distinguishes between local and non‐
local searchers to determine to what extent their residential search process differs.
The study also explores the usefulness of a new method to study the residential
search process: a diary approach with both qualitative and quantitative elements.
The chapter presents the results of a small‐scale study of ten respondents looking for
houses in the northern Netherlands. The purpose is to offer additional insights to the
quantitative approach used in the other chapters about the process preceding a
move. In this chapter no distinction between less‐popular and popular areas is
made. The objective was to find out how people find places to live in rural areas in
general, not to compare the search process in less‐popular and popular areas.
Finally, Chapter 6 summarises and discusses the main findings drawn from the
analyses in Chapters 2 to 5. Furthermore, this chapter also considers policy
implications and suggestions for further research.
References
Andersen HS. 2011. Explanations for long‐distance counter‐urban migration into
fringe areas in Denmark. Population, Space and Place 17: 627‐641.
Appleton J. 1975. The Experience of Landscape. Wiley: London.
Argent N, Smailes P, Griffin T. 2007. The amenity complex: towards a framework for
analysing and predicting the emergence of a multifunctional countryside in
Australia. Geographical Research 45: 217‐232.
Beale C. 1975. The revival of population growth in non‐metropolitan America. United
States Department of Agriculture, Economic Research Service, ERS 605.
Berry B (ed.). 1976. Urbanization and Counterurbanization.Sage: Beverly Hills,
California.
Bettman JR. 1979. An Information Processing Theory of Consumer Choice. Addison‐
Wesley: Reading.
20
Bollman RD, Briggs B. 1992. Rural and small town Canada: an overview. In Rural
and Small Town Canada, Bollman RD (ed.). Thompson Educational Publishing:
Toronto.
Bolton N, Chalkley B. 1990. The rural population turnround: a case‐study of North
Devon. Journal of Rural Studies 6: 29‐43.
Boyle P, Halfacree K. 1998. Migration into Rural Areas: Theories and Issues. John Wiley
& Sons: Chichester.
Champion AG (ed.). 1989. Counterurbanization: the Changing Pace and Nature of
Population Deconcentration. Edward Arnold: London.
Coolen H, Hoekstra J. 2001. Values as determinants of preferences for housing
attributes. Journal of Housing and the Built Environment 16: 285‐306.
Coolen H, Boelhouwer P, Van Driel K. 2002. Values and goals as determinants of
intended tenure choice. Journal of Housing and the Built Environment 17: 215‐236.
Deller, SC, Tsai, T, Marcouiller, DW, English, DBK. 2001. The role of amenities and
quality of life in rural economic growth. American Journal of Agricultural
Economics 83: 352‐365.
Elbersen B. 2001. Nature on the Doorstep: the Relationship between Protected Natural
Areas and Residential Activity in the European Countryside. Wageningen: Alterra.
Feijten, P, Hooimeijer, P, Mulder, CH. 2008. Residential experience and residential
environment choice over the life‐course. Urban Studies 45: 141‐162.
Fitchen JM. 1995. Spatial redistribution of poverty through migration of poor people
to depressed rural communities. Rural Sociology 60: 181‐201.
Foulkes M, Newbold KB. 2008. Poverty catchments: migration, residential mobility,
and population turnover in impoverished rural Illinois communities. Rural
Sociology 73: 440‐462.
Gkartzios M, Scott M. 2009. Residential mobilities and house building in rural
Ireland: evidence from three case studies. Sociologia Ruralis 50: 64‐84.
Grimsrud GM. 2011. How well does the ‘counter‐urbanisation story’ travel to other
countries? The case of Norway. Population, Space and Place 17: 642‐655
Haartsen T, Huigen PPP, Groote P. 2003. Rural areas in the Netherlands. Tijdschrift
voor Economische en Sociale Geografie 94: 129‐136.
21
Haartsen T, Venhorst V. 2010. Planning for decline: anticipating on population
decline in the Netherlands. Tijdschrift voor Economische en Sociale Geografie 101:
218‐227.
Haartsen T. 2012. Kennisagenda Krimp. Kennisnetwerk Krimp Noord‐Nederland.
Halfacree K. 1994. The importance of ‘the rural’ in the constitution of
counterurbanization: evidence from England in the 1980s. Sociologia Ruralis 34:
164‐189.
Halfacree K. 2001. Constructing the object: taxonomic practices,
‘counterurbanisation’ and positioning marginal rural settlement. International
Journal of Population Geography 7: 395‐411.
Halfacree K. 2008. To revitalise counterurbanization research? Recognising an
international and fuller picture. Population, Space and Place 14: 479‐495.
Halfacree K. 2012. Heterolocal identities? Counter‐urbanisation, second homes, and
rural consumption in the era of mobilities. Population, Space and Place 18: 209‐
224.
Halliday J, Coombes M. 1995. In search of counterurbanisation: some evidence from
Devon on the relationship between patterns of migration and motivation.
Journal of Rural Studies 11: 433‐446.
Heijs W, Carton M, Smeets J, Van Gemert A. 2009. The labyrinth of life‐styles. Journal
of Housing and the Built Environment 24: 347‐356.
Heijs W, Van Deursen A, Leussink M, Smeets J. 2011. Re‐searching the labyrinth of
life‐styles. Journal of Housing and the Built Environment 26: 411‐425.
Heins S. 2002. Rurale Woonmilieus in Stad en Land: Plattelandsbeelden, Vraag naar en
Aanbod van Rurale Woonmilieus. Eburon: Delft.
Hoggart, K. 2007. The diluted working classes of rural England and Wales. Journal of
Rural Studies 23: 305‐317.
Jansen SJT. 2011. Lifestyle method. In The Measurement and Analysis of Housing
Preference and Choice, Jansen SJT, Coolen HCCH, Goetgeluk RW (eds.).
Springer: Dordrecht; 177‐202.
Johnson KM, Beale CL. 1994. The recent revival of widespread population growth in
nonmetropolitan areas of the United States. Rural Sociology 59: 655‐667.
22
Kersloot J, Kauko T. 2004. Measurement of housing preferences: a comparison of
research activity in the Netherlands and Finland. Nordic Journal of Surveying and
Real Estate Research 1: 144‐163.
Markantoni M. 2012. Side activities by non‐farmers. PhD Thesis Faculty of Spatial
Sciences: University of Groningen.
Marlet G. 2009. De Aantrekkelijke Stad. VOC Uitgevers: Nijmegen.
McGranahan DA. 2008. Landscape influence on recent rural migration in the U.S.
Landscape and Urban Planning 85: 228‐240.
Milbourne P. 2007. Re‐populating rural studies: migrations, movements and
mobilities. Journal of Rural Studies 23: 381‐386.
Rokeach MJ. 1973. The Nature of Human Values. Free Press: New York.
Shumway JM, Otterstrom SM. 2001. Spatial patterns of migration and income
change in the mountain west: the dominance of service‐based, amenity‐rich
counties. Professional Geographer 53: 492‐502.
Smith D. 2007. The changing faces of rural populations:‘“re‐fixing” the gaze’ or ‘eyes
wide shut’? Journal of Rural Studies 23: 275–282.
Smith, DP, Phillips, D.A. 2001. Socio‐cultural representations of greentrified Pennine
rurality. Journal of Rural Studies 17: 457‐469.
Steenbekkers A, Simon C, Vermeij L, Spreeuwers W. 2008. Het Platteland van Alle
Nederlanders: Hoe Nederlanders het Platteland Zien en Gebruiken. Sociaal en
Cultureel Planbureau: Den Haag.
Stockdale A. 2006. Migration: Pre‐requisite for rural economic regeneration? Journal
of Rural Studies 22: 354‐366.
Ulrich RS. 1986. Human responses to vegetations and landscapes. Landscape and
Urban Planning 13: 29‐44.
Van Dam F. 2000. Revealed and stated preferences for rural living: evidence from
the Netherlands. In Claiming Rural Identities: Dynamics, Contexts, Policies,
Haartsen T, Groote P, Huigen PPP (eds.). Van Gorcum: Assen; 80‐91.
Van Dam F, Heins S, Elbersen BS. 2002. Lay discourses of the rural and stated and
revealed preferences for rural living: some evidence of the existence of a rural
idyll in the Netherlands. Journal of Rural Studies 18: 461‐476.
23
Van den Berg AE, Koole SL, Van der Wulp NY. 1998. Environmental preference and
restoration: (how) are they related? Journal of Environmental Psychology 23: 135‐
146.
Walmsley DJ, Epps WR, Duncan CJ. 1998. Migration to the New South Wales North
Coast 1986‐1991: Lifestyle motivated counterurbanisation. Geoforum 29: 105‐
118.
Westlund H. 2002. An unplanned green wave: settlement patterns in Sweden during
the 1990s. Environment and Planning A 34: 1395‐1410.
Woods M. 2005. Rural Geography. Sage Publications: London.
Woods M. 2011. Rural. Routledge: Oxon/New York.
Yu K. 1995. Cultural variations in landscape preference: comparisons among
Chinese subgroups and Western design experts. Landscape and Urban Planning
32: 107‐126.
24
25
2. More than counter‐urbanisation: Migration to popular and
less‐popular rural areas in the Netherlands2
Abstract
Migration into rural areas in Western countries is often explained by the pull of the rural
idyll for urban, middle‐class migrants. While previous research has shown that this counter‐
urbanisation model is insufficient to explain rural immigration in sparsely populated
countries, this paper shows that also within core regions, more diverse conceptualisations of
migration into rural areas are required. This is achieved by distinguishing popular, average
and less‐popular rural living areas in the northern Netherlands, based on average house
prices, and by analyzing the migration flows to these areas. Data from Housing Research of
the Netherlands demonstrate that popular rural areas attract more highly‐educated people
and people moving from urban areas, compared to less‐popular and average rural areas.
For movers to less‐popular areas, being near to family and friends is more important. The
characteristics of the movers to popular rural areas fit very well with the counter‐
urbanisation story. Less‐popular rural areas in the Netherlands share personal reasons as an
important motivation for in‐migration with more remote rural areas in Europe. This
indicates that conceptualisations of periphery and remoteness have to be considered within
the local, regional and national context. Research into rural population change in both core
regions and sparsely populated countries should consider these different contexts to be able
to acknowledge the variety in the way amenities and peripherality are perceived by
different groups of people.
2.1 Introduction
Increasingly, the idea that counter‐urbanisation – in the sense of the middle‐class,
urban dream of escaping to the countryside – can explain all rural population
change in Western society rural areas has been challenged (Milbourne, 2007;
2 This chapter is reprinted from: Bijker, RA, Haartsen T. 2012. More than counter‐urbanisation:
Migration to popular and less‐popular rural areas in the Netherlands. Population, Space and Place 18:
643‐657. Copyright © 2012 John Wiley & Sons, Ltd.
26
Halfacree, 2008). In an attempt to bridge the gap between the ‘counter‐urbanisation
story’ and the lack of empirical evidence for this in Norway, Grimsrud (2011) calls
for caution in applying conceptualisations of rural in‐migration originating in the
core regions of Europe and America to more sparsely populated countries. In this
paper we want to broaden the discussion by stating that within the core regions of
Europe, migration patterns in rural areas can also not sufficiently be explained by
the prevailing counter‐urbanisation models alone. This connects to what Milbourne
(2007) points out to be the spatial complexity of rural population change, which
exists not only between, but also within countries.
One of the aspects of this spatial complexity is the finding that rural areas
differ in their popularity for rural living (Woods, 2005). Spatial characteristics or
amenities such as landscape, location relative to coasts or beaches, and distance from
metropolitan centres influence the popularity of rural areas for migration purposes
(Argent et al., 2007; Bollman and Briggs, 1992; Hugo and Smailes, 1985; Johnson and
Beale, 1994; McGranahan, 2008). In addition, perceived lifestyle and employment
opportunities play a role (Johnson and Beale, 1994).
Rural migration is also diverse in terms of migrant characteristics (Andersen,
2009; Guimond and Simard, 2010; Hugo and Bell, 1998; Lindgren, 2003; Mitchell et
al., 2004; Philips, 1998; Stockdale, 2006b) and motives. Some migrants are motivated
by a desire for a rural lifestyle and residence, while others move to the countryside
for pragmatic reasons, such as lower housing costs or work opportunities (Fitchen,
1995; Gkartzios and Scott, 2009; Halfacree, 1994; Kasimis et al., 2003).
While it seems reasonable to assume that the characteristics of rural areas relate
to the characteristics and motives of migrants, this relatedness has not been studied
often. Argent et al. (2007, p. 218) note that ‘perceived rural amenity is clearly an
anthropocentric concept, relating to the subjective interpretation of aspects of the
physical and cultural environment in terms of their creation of a “pleasant” setting’.
It depends on a wide variety of human factors and preferences, including age,
fitness and socioeconomic status. Hjort and Malmberg (2006) also relate the diversity
of the countryside, ‘a number of contrasting countrysides’, to the many different
meanings and motives moving to rural areas may involve. Studies in more sparsely
27
populated countries like Sweden and Scotland provide some clues suggesting that
different rural areas do indeed attract different types of movers with different
motivations and preferences (Hjort and Malmberg, 2006; Stockdale, 2006a). To what
extent these findings can be translated to the more densely populated countries in
the core regions in Europe is less clear.
This paper investigates whether different types of rural areas also attract
specific migrants with specific motivations within core regions. We differentiate
rural areas by focusing on differences in their popularity. In our analysis of the
characteristics and motivations of people who migrate into these rural areas, we take
into account migration flows from both urban and rural areas, and local and long‐
distance migration flows. We thus try to address Milbourne’s (2007:385) appeal for
thinking ‘more critically about the broad range of movements and mobilities that are
being played out in rural spaces’. He states that ‘rural researchers have been
preoccupied with longer distance movements and with migrations from urban to
rural places. While such movements provide an important research subject,
particularly if the objective is to identify differences, tensions and conflicts between
new and established fractions of the rural population, they do run the risk of
marginalising other types of movement’ (ibid.).
Our research area is located in the northern Netherlands. In comparison with
other European countries, the Netherlands is densely populated and urbanised.
According to the OECD standard, there are no predominantly rural areas left in the
Netherlands (Terluin et al., 2005). In the Dutch context, however, a considerable part
of the country is considered rural by official bodies such as Statistics Netherlands, as
well as by the Dutch people (Haartsen, 2002; Haartsen et al., 2003). We focus on the
northern part of the country, consisting of the provinces of Friesland, Groningen and
Drenthe (see Figure 3), because the North can be considered to be the most rural part
of the Netherlands, based both on address density and the perceptions of the Dutch
people (Haartsen et al., 2003; Heins, 2002). In addition, the North contains a large
variety of rural areas in terms of characteristics such as landscape, presence of
natural areas, accessibility, building style of houses and history.
28
This paper compares migration patterns to different types of rural areas, based
on the characteristics and motives of the people migrating there. After some
theoretical discussion, the analysis commences by describing migration figures to
rural areas in the northern Netherlands in general. We then distinguish popular and
less‐popular rural living areas. We attempt to characterise them and provide an
overview of migration flows to the different types of rural areas. The characteristics
and motivations of those migrating are compared. To do so, we have used data from
Statistics Netherlands and from Housing Research of the Netherlands. Based on the
results from these analyses, we try to prove that looking at differences in rural areas
and at differences in the people they attract, can provide a more complete picture of
migration patterns in core regions.
2.2 Popularity of rural areas
In the literature it is acknowledged that migration flows to rural areas are spread
unevenly. According to Selwood et al. (1995), no across‐the‐board repopulation of
‘the’ rural is occurring, but population flows are focused on particular places with
scenic or Arcadian ambience. Accordingly, the popularity of rural areas for living
depends, at least partly, on the amenity of their local physical, situational and
economic environments (Argent et al., 2007).
The local physical environment is often expressed in terms of the different
landscape ingredients of areas. Landscape has been found to influence migration.
People are most drawn to areas with a mix of forest and open land, water area,
topographical variation and relatively little cropland (McGranahan, 2008). These
results correspond with the more general findings on landscape preferences, in that
people strongly prefer park or savannah‐like landscapes with scattered trees or
small stands, open vistas, an even and uniform ground texture, and a water source
(Ulrich, 1986). These preferences appear to be largely independent of culture, with
the exception of people dependent on other landscapes for their livelihood, such as
farmers and foresters (Van den Berg et al., 1998; Yu, 1995). Preferences for semi‐open
landscapes can be explained by the so‐called prospect‐refuge theory. According to
29
Appleton (1975), the ability to see (prospect) without being seen (refuge) increases
perceived safety, which in turn increases the aesthetic pleasure experienced.
In addition to landscape amenities, situational aspects of rural areas are
important for the perceived attractiveness of these rural areas for living. Gkartzios
and Scott (2009) state that in Ireland, impressive rural repopulation is found in
accessible rural areas, along the road and railways, while the population is declining
in less accessible rural areas, particularly in the west. Transport availability is not the
only factor determining the accessibility of rural areas, as proximity to cities also
plays a role. In the US, Canada and Sweden, population growth is concentrated in
the rural districts closest to metropolitan and regional urban centres, while the more
remote rural regions suffer stagnation or decline (Bollman and Briggs, 1992; Johnson
and Beale, 1994; Westlund, 2002).
With reference to the economic environment, Johnson and Beale (1994) found
that rural areas that are centres of recreation grow quickest. This characteristic also
comes to the fore in the amenity index for Australian rural areas developed by
Argent et al. (2007), which is based on a broad range of biophysical, climatic,
historic, geographic and economic amenity indicators. They found that immigration
rates correlate with beach distance, irrigation water resources and employment in
recreational and related services. Argent et al. (2007) see employment in recreational
and related services as a measure of the attractiveness of a rural area to live in, with
areas with significant natural and artificial recreational attractions scoring high.
They also point out that many in‐migrants develop a familiarity with a future abode
through recreational visits. There is thus a connection between an area’s attracting
short‐term visits and longer‐term moves.
The abovementioned features seem to be connected with the popularity of
rural areas as places to live in several countries. Later in this article we will use them
to describe the popular and less‐popular rural areas we distinguish in the northern
Netherlands.
30
2.3 Different people, different motives?
Research into rural migration is dominated by the concept of counter‐urbanisation
(Champion, 1989; Boyle and Halfacree, 1998), which is often associated with the
movement of middle‐class families from cities towards ‘what is perceived to be an
idyllic rural setting’ (Halfacree, 2008:479). The rural idyll refers to a very positive
image of the countryside comprising many aspects of rural lifestyle, community and
landscape (Ilbery, 1998). Research into rural population change has focused on
whether the ‘rural idyll’ exists and its importance to rural in‐migration decisions
(Benson and O’Reilly, 2009; Hjort and Malmberg, 2006; Van Dam et al., 2002).
People’s positive images of the rural and its attractiveness may also be related to a
specific part of a country, for example the area where a person either grew up,
purchased a second home, holidayed, or a specific landscape described in art, film or
literature (Hjort and Malmberg, 2006; Selwood et al., 1996).
However, other studies found the characteristics of people moving to the
countryside to be far more diverse than the classical counter‐urbanisation model
suggests: for example, Halfacree (2008) notes that counter‐urbanisers are not only
middle class, but also working class. Moreover, rural migration is also more
complicated than just movements over longer distance in the urban‐to‐rural
direction (Milbourne, 2007). In their case studies in rural Ireland, Gkartzios and Scott
(2009) found a mixture of residential movements, including long‐distance moves
from bigger, similar or even smaller settlements, but also a significant number of
local residential movements within the same area.
With regard to the motivations of movers to rural areas, rural idyllic motives
related to the social and physical characteristics of the rural destination do not
explain all rural moves either. Migrants can have other motives, related to housing
characteristics, lower housing costs, work‐related reasons and living close to family
and friends (Fitchen, 1995; Gkartzios and Scott, 2009; Grimsrud, 2011; Halfacree,
1994; Halliday and Coombes, 1995; Kasimis et al., 2003; Marshall et al., 2004;
Steenbekkers et al., 2008; Walmsley et al., 1998).
31
2.4 Different areas, different people?
In an international context some research has been done directly comparing the
features of rural areas and the characteristics and motives of people migrating to
different types of rural areas. Referring to the local physical environment, the first of
the three characteristics that influence the popularity of rural areas we discussed
earlier, Elbersen (2001) found that rural areas located close to protected natural areas
attract more early‐retirees, footloose and middle‐class households, and households
moving from urban areas, compared to other rural areas in the Netherlands. In
addition, Amcoff (2000, as cited in Hjort and Malmberg, 2006) observes that Swedish
rural locations most associated with the rural idyll, tend to attract people with
university‐level education. Apparently, for areas with a rural idyllic landscape, the
stereotypical counter‐urbanisation model does apply. To what extent this is true for
less idyllic rural areas, which are also present within core regions, is not clear.
The second characteristic refers to the situational aspects of rural areas. In their
study of characteristics of rural migrants in Sweden, Hjort and Malmberg (2006)
investigated to what extent rural areas located further from urban centres attract
different migrants than areas located closer by. They found that people aged 19–40,
high‐income earners, people with university‐level education and singles were more
likely to move to the peri‐urban countryside, while people older than 61 were more
likely to end up in the remote countryside. Stockdale (2006a), who examines
migration to peripheral and depopulating areas of rural Scotland, adds that most
migrants moved over short distances, often from elsewhere in the study region. In‐
migration also consisted of urban‐to‐rural movements as well as rural‐to‐rural
movements. Contrary to much of the counter‐urbanisation literature on the
repopulation of rural areas (Boyle and Halfacree, 1998, Stockdale et al., 2000),
quality‐of‐life considerations hardly played a role in the Scottish periphery. Based
on this, Stockdale concludes that the motivations for in‐migration to depopulating
and repopulating areas apparently differ markedly. In the depopulating peripheral
areas, personal reasons – relating to marriage and divorce, for example – are the
most important, followed by employment considerations. Grimsrud (2011) found a
32
similar pattern in Norway, where movers to remotely located rural areas are often
motivated by family ties to the area and job‐related reasons. Movers to the urban
fringe, which she defines as rural areas at commuting distance from a big city, better
fit the counter‐urbanisation narrative, with housing and social and physical
amenities appearing as important motivations. Since we intend to determine
whether spatial complexity of rural population change also occurs in rural areas in
core regions, we think it is important to compare our results with what we know
about migration patterns to both urban fringe areas and remote rural areas in more
sparsely populated countries in Europe.
The third characteristic influencing the popularity of rural areas is the
economic environment. Shumway and Otterstrom (2001) compare migration flows
to rural areas with service‐based economies and high levels of natural amenities in
the U.S., to areas with ‘old’ economic structures, such as farming or mining. They
find that in‐migrants to counties with service‐based economies have the highest
incomes when compared to migrants to the other areas.
2.5 Facts and figures of rural migration in the northern Netherlands
Before we start our analysis of popular and less‐popular rural areas, and the
similarities and differences regarding in‐migrants and their motives, we will provide
an overview of the population and migration figures for the northern Netherlands in
general, and for the rural parts of the North more specifically. In total, around
1,700,000 people live in the region, or about 10% of the total Dutch population. The
northern Netherlands consists of three provinces and 68 municipalities. Based on
Statistics Netherlands’s national standard for urbanisation, the so‐called address
density, 58 of these municipalities are rural, i.e. municipalities with an average of
less than 1,000 addresses per square kilometre. Sixty‐five percent of the northern
population lives in these rural municipalities.
As Figure 1 shows, the northern net migration is slightly negative. In the
period 2003–2007, a five‐year annual average of 22,326 people leave the North for
elsewhere in the Netherlands, while 21,913 people arrive. Of these in‐migrants from
33
elsewhere in the Netherlands, 54% find a place to live in the countryside. Around a
third of the in‐migrants to rural municipalities originate from other Dutch rural
areas. Sixty‐eight percent (7,967 people) can be considered counter‐urbanists,
moving from an urban municipality elsewhere to a rural municipality in the North.
This is more than the overall Dutch average: according to Steenbekkers et al. (2008),
in 2006, 57% of the people who moved to Dutch rural areas came from an urban
area.
Figure 1. Yearly migration from and to the North, from elsewhere in the Netherlands (5 year
average, 2003‐2007)
In addition to migration from other parts of the Netherlands, migration flows
into rural northern areas also consist of internal migration. Figure 2 shows that the
five‐year annual average of migration between northern Netherlands municipalities
is 51,831. Sixty‐one percent of these migrants move into a rural municipality, of
which almost two‐thirds (19,236) originate from another rural municipality and
around a third from urban areas in the North. However, a significant proportion
(29%; 15,049) of the migrants move from the countryside to urban municipalities.
This overview of the migration flows towards and within the northern
Netherlands supports the view of Milbourne (2007), that rural population dynamics
consist of a broad range of movements and mobilities. In addition to urban‐rural
34
migration, there is a considerable migration flow between rural areas within the
northern Netherlands, which is even larger than the urban‐rural migration. This
range of movements raises the question of whether variation also exists within these
rural areas in terms of migration numbers and the characteristics and motives of the
migrants.
Figure 2. Yearly internal migration in the North of the Netherlands, between municipalities
(5 year average, 2003‐2007)
2.6 Defining popular and less‐popular rural areas in the northern Netherlands
We distinguish popular and less‐popular rural areas based on average house prices.
House prices are an indication of what people are willing to pay for houses and their
environments (Luttik, 2000; Visser and Van Dam, 2006), thus reflecting the value
buyers attach to houses and their surroundings. Therefore we consider the average
house prices in an area as an indicator of the popularity of an area for living. This
average house price incorporates the residential environment as well as housing
characteristics such as size and quality.
The use of house prices instead of in‐migration numbers, which are often used
in studies of rural areas’ popularity (Argent et al., 2007; McGranahan, 2008), is more
appropriate for the Dutch context. In his research into Dutch cities’ attractiveness
based on differences in house prices, Marlet (2009) shows that no relationship exists
35
between cities’ attractiveness and population growth in the Netherlands. It appears
that population growth can mainly be explained by government development
policy. New housing is developed in a limited number of areas, while in others,
building houses is very much restricted, especially in rural areas, both in open
country and in villages (Van Dam et al., 2002). These restrictions prevent the Dutch
housing market from reacting quickly to changes in demand for new houses,
causing a very low elasticity of supply. This leads to price increases instead of
population growth when there is increasing demand (Marlet, 2009). Therefore, the
popularity of areas can be investigated using differences in house prices. Glaeser
and Shapiro (2003) show that the relationship between amenities such as a
temperate climate and growth has also weakened in the United States, something
they explain by the anti‐growth movement in California limiting new housing
supply. McGranahan (2008) also observes that in a US context, housing‐supply
constraints such as land‐use regulations could be mitigating the relationship
between landscape preferences and migration in rural areas.
The average housing prices per municipality for the rural municipalities in the
northern Netherlands varied from EUR 142,600 to EUR 340,000 in 2008. We
performed a cluster analysis of the average housing prices per municipality for the
rural municipalities in the northern Netherlands. The outcome consisted of four
clusters, one containing only one case (with the highest average price). We combined
this one‐case cluster with the cluster with the second‐highest price average. This
resulted in the following three clusters:
1. less‐popular rural areas (n=18), lowest housing prices (cluster average price
EUR 174,828)
2. average rural areas (n=26), average housing prices (cluster average price
EUR 212,731)
3. popular rural areas (n=10), highest housing prices (cluster average price
EUR 273,170).
36
Figure 3 shows that the less‐popular areas are located along the northern and
eastern borders of the northern Netherlands. More than half of these less‐popular
rural municipalities border the Wadden Sea, an intertidal zone with mud flats and
wetlands. The coastline consists of a typical Dutch dike separating the mud flats
from agricultural land, perhaps explaining why the northern Netherlands’ coastline
is less popular than the Australian coast, with its sandy beaches and cliffs (Costello,
2007).
Figure 3. Three types of rural areas in the North of the Netherlands
To characterise the three types of areas and to describe the similarities and
differences we calculated their mean scores for several characteristics. We used
those characteristics which in an international context have been shown to differ
between popular and less‐popular rural areas: landscape, accessibility and
employment in recreation‐related sectors (e.g. Argent et al., 2007; Johnson and Beale,
1994; McGranahan, 2008). We also chose a number of sociocultural characteristics,
socioeconomic characteristics and characteristics related to future population
development. The results are presented in Table 1.
37
Table 1. Mean scores for different characteristics, per type of rural area (one‐way analysis of
variance) Less‐popular
rural areas
(n=18)
Average
rural areas
(n=26)
Popular
rural areas
(n=10)
Total2
Landscape characteristics
Share in land use (2003) Agriculture 82% 85% 78% 82%
Nature 6% 6% 13% 7%*
Recreation 1% 1% 2% 1%*
Water 0.21% 0.68% 0.14% 0.13%
Urban area 7% 5% 5% 5%
Dominant type of soil (%
of municipalities in
cluster)
Marine clay 67% 42% 10% 44%*
Former peat 22% 31% 30% 28%
Sand 11% 27% 60% 28%*
Type of agriculture (% of
total agricultural area,
2005)
Arable farming 49% 20% 10% 27%*
Horticulture 1% 0% 0% 0%
Grazing livestock 40% 74% 84% 64%*
Non‐grazing
livestock
0% 0% 0% 0%
Combinations 9% 5% 5% 6%*
Accessibility1
Travel distance to nearest urban centre in the
northern Netherlands (in minutes)
22 24 24 24
Travel distance to nearest urban centre in the
central Netherlands (in minutes)
109 93 84 97*
Employment in recreation related services
Proportion employment in hotel and catering
industry (2006)
3% 5% 6% 4%*
Sociocultural characteristics
Dimensions of
Dutch regional
culture 1997‐
2003 (after Brons,
2005;2006)
Post‐materialism 0.094 ‐0.365 ‐0.281 ‐0.210*
Protestant conservatism 0.099 0.085 ‐0.110 0.0538
Classic individualism 0.170 0,449 0.707 0.411*
Egalitarian anti‐
conservatism
1.435 1.165 1.244 1.270
Dissatisfaction ‐0.512 ‐0.914 ‐0.998 ‐0.795
Socioeconomic characteristic
Average income per household per
municipality (in euros, 2006)
19,206 20,150 22,040 20,185*
Future population development
Expected population development (2007‐
2025)
‐9.321 ‐4.957 ‐4.336 ‐6.297
Expected household development (2007‐2025) ‐4.142 4.782 6.682 2.159*
Source: Statistics Netherlands, except accessibility
1 Measured from the town hall of the specific rural municipality to the city hall in the nearest provincial capital
city or the nearest of the two cities in the centre of the country, as predicted by a website on travel routes.
2 * p<0.05.
38
We used three variables to compare the landscape of the three types of rural
areas. The first is land use. Table 1 shows that popular rural areas have a
significantly larger share of the nature and recreation land‐use categories. This
corresponds with Elbersen’s findings (2001) that the vicinity of natural areas attracts
new residents. The second and third – ‘dominant type of soil’ and ‘type of
agriculture’ – are measures of the type of landscape of the specific rural areas. In the
northern Netherlands, two dominant landscape types exist. The first is the open,
young marine clay and former peat landscape, where arable farming and large‐scale
dairy farming prevails. The second is the semi‐open sand landscape, where small‐
scale mixed and dairy farming is dominant (Haartsen and Strijker, 2010). Less‐
popular areas are characterised by a large share of marine clay soils and a large
share of arable farming. Popular rural areas are dominated by sand landscapes and
grazing livestock farming. This is consistent with the findings of Ulrich (1986) and
McGranahan (2008), showing that people prefer areas with a mix of forest and open
land and relatively little cropland.
The second characteristic is related to accessibility, measured as the distance to
the nearest provincial capital city (including Zwolle, the capital of the neighbouring
province Overijssel), and to the nearest city in the central part of the Netherlands
(Amsterdam or Utrecht, see Figure 3), measured in travel time in minutes by car.
Mean travel distances to the nearest city in the central Netherlands vary significantly
for the three types of rural areas. Popular rural areas are located closer to the centre
of the country.
Following Argent et al. (2007), we compare the proportion of employment in
the hotel and catering industry per municipality for the three types of area. This is
the best available proxy variable for measuring employment in recreation and
related services. The proportion differs significantly; the share is highest in popular
rural areas and lowest in the less‐popular areas.
Although it is hard to operationalise social features of rural areas (Argent et al.,
2007), we tried to determine whether the three types of rural areas differ for five
specific dimensions of regional culture. Brons (2005; 2006) defined these dimensions
when he developed a dataset based on quantitative measurements of the culture of
39
Dutch municipalities. The dimensions of regional culture were developed from
various indicators using factor analysis. Less‐popular rural areas appear to have a
regional culture that contains significantly more post‐materialism. This dimension is
related to anti‐conservatism, feminist values and self‐expressive individualism. It
consists, for example, of the number of one‐person households, progressive parties
voters and the number of children with unmarried parents. However, high scores on
these variables can also be interpreted less positively. A large number of one‐person
households and children born outside of marriage could also indicate some degree
of social problems there. The regional culture of popular rural areas can be
characterised by significantly more classical individualism, i.e. personal freedom,
materialism and egoism. Its indicators are the percentage of votes for liberal parties,
and the postponement of marriage and having children.
The socioeconomic characteristics are measured in terms of the average income
per household per municipality in 2006. In less‐popular rural areas, households have
a significantly lower average income than those in average and popular rural areas.
Popular areas have households with the highest average income.
Finally, we believe that figures on the expected growth of both the population
and the number of households in 2025 (compared to 2007) are relevant to
characterise the popularity of the different types of rural areas. In the Netherlands,
between 2010 and 2040, low and negative growth are expected for a large number of
municipalities. These areas are mostly located on the periphery, including the north.
Rural areas are expected to experience population decline more seriously and earlier
than urban areas (Haartsen and Venhorst, 2010). The expected growth in the number
of households is also relevant in housing policy, since the Netherlands has been
experiencing a trend of decreasing household size, which implies that even where
there is population decline, the number of households can increase. As Table 1
shows, the expected population decline does not significantly differ for the different
types of rural areas, although less‐popular areas do have a higher expected
population decline than average and popular rural areas. However, less‐popular
areas are expected to experience a decline in the number of households, while the
number of households is expected to increase in average and popular areas.
40
Therefore, regarding population and household developments, less‐popular areas
can be considered as areas of decline.
To summarise, we found that less‐popular rural areas are located along the
northern and eastern borders of the North of the Netherlands. They can be
characterised as having a relatively open, marine clay landscape with a large share
of large‐scale arable farmland and relatively less natural and recreational areas.
They are relatively distant from the centre of the Netherlands, and have a relatively
low share of hotel and catering industry employment. The people who inhabit these
areas have a lower average income than those in the average and popular rural
areas, and the regional culture is more dominated by ‘post‐materialism’. The
variables that indicate this cultural dimension could also hint at some degree of
social difficulties. Population and household decline are expected in the less‐popular
areas.
Popular rural areas can be characterised as having a more semi‐open landscape
with predominantly grazing livestock farming and a relatively large share of natural
areas. These areas’ inhabitants have a relatively high average income and the areas
have a relatively high share of hotel and catering industry employment. Classical
individualism is a dimension of regional culture that prevails in these areas. Popular
rural areas have an expected household growth and are the least distant from the
centre of the Netherlands. The average rural areas score between the other rural
areas for most characteristics.
2.7 Differences in migration patterns to popular and less‐popular rural areas
Having defined and characterised less‐popular, average and popular rural areas in
the North of the Netherlands, we will now focus on the migration flows to and from
these areas. The data we use were obtained from Statistics Netherlands.
41
Table 2. Mean net migration rate and in‐migrants, per type of rural area (one‐way analysis
of variance)
Less‐popular
rural areas (n=18)
Average rural
areas (n=26)
Popular rural
areas (n=10)
Total
Five‐year average annual net
migration rate 2003‐2007
‐3.7 ‐2.3 ‐1.5 ‐2.6
Five‐year average annual in‐
migrants per 1,000 inhabitants
2003‐2007
40 41 42 41
Source: Statistics Netherlands
As Table 2 shows, all three types of rural areas show a negative net migration
for the period 2003–2007, and the size of the net migration rate does not differ
significantly between the three types of areas as well. This means that more people
leave the area than move to it, even in popular areas. The average number of in‐
migrants per 1,000 inhabitants does not differ significantly for the three types of
areas in the same period. Popular rural areas do not attract significantly more
migrants than less‐popular rural areas either.
Although the three types of rural areas show no differences in overall
migration figures, it could well be that they do attract different groups of migrants.
We analysed this for the origin and the age of migrants (see Table 3). Popular rural
areas attract a higher share of movers from outside the northern Netherlands than
the less‐popular and average rural areas. Popular areas also attract the highest share
of movers from urban areas, less‐popular areas receive relatively the least urban‐to‐
rural migrants. The age distribution shows that movers to less‐popular and average
areas are in general younger than movers to popular areas.
42
Table 3. Characteristics of in‐migrants for three types of rural areas in the North of the
Netherlands.
To
Less‐
popular
rural areas
(n=18)
Average
rural
areas
(n=26)
Popular
rural areas
(n=10)
Total
Moved from:
North of the
Netherlands
74% 73% 70% 73%
Elsewhere in the
Netherlands
26% 27% 30% 27%
Total 100% 100% 100% 100%
Moved from:
Urban areas 42% 46% 55% 47%
Rural areas 58% 54% 45% 53%
Total 100% 100% 100% 100%
Age group:
0‐20 26% 24% 23% 24%
20‐30 27% 27% 23% 26%
30‐40 18% 19% 21% 19%
40‐50 12% 12% 12% 12%
50‐65 10% 11% 13% 11%
65+ 7% 7% 8% 7%
Total 100% 100% 100% 100%
Source: Statistics Netherlands
When we focus on migration within the North we can also analyse the
migration flows between the different types of rural areas and urban areas. Table 4
shows that less‐popular areas receive the largest share of people migrating from
other less‐popular areas, while they receive relatively fewer migrants from popular
areas. Less‐popular rural areas attract the lowest percentage of urban in‐migrants,
whereas popular rural areas receive the highest share of urban in‐migrants. More
than 50% of the migrants to these rural areas come from an urban municipality.
Popular areas welcome relatively few people migrating from less‐popular and
average areas.
43
It seems that a considerable migration flow exists between urban areas and
popular rural areas, while the migration flow between less‐popular and average
rural areas, and popular rural areas is very limited. This corresponds with Hjort and
Malmberg’s finding (2006) that population trends in the Swedish countryside are
highly dependent on migration to and from the cities, while the exchange between
the peri‐urban and remote countryside is very limited.
Table 4. Migration within the North of the Netherlands towards the different types of rural
areas.
To
From: Less‐popular
rural areas
(n=18)
Average
rural areas
(n=26)
Popular
rural areas
(n=10)
Total
Less‐popular rural
areas
28% 21% 10% 21%
Average rural areas 34% 31% 21% 30%
Popular rural areas 6% 10% 17% 10%
Urban
municipalities
32% 39% 51% 39%
100% 100% 100% 100%
Source: Statistics Netherlands
2.8 Characteristics and motivations of in‐migrants: less‐popular, average and
popular rural areas compared
The Statistics Netherlands data permitted an overview of migration numbers and
the origin and age composition of migrants to rural areas in the northern
Netherlands. To complement the above analysis with a comparison of the
characteristics and motivations of migrants to the different types of rural areas, we
analysed data from Housing Research of the Netherlands (WoON 2009), which
investigates housing preferences and circumstances in the Netherlands every three
years. The total sample consists of 78,000 respondents. For our analysis, we selected
respondents living in rural municipalities in the northern Netherlands from the
national sample. Rural municipalities are again defined by an address density of less
44
than 1,000 addresses per square kilometre. From this group of rural respondents, we
selected people who had moved to their current address from another municipality
in the previous two years. This definition of migrants was chosen because the
questions related to the decision to move in the questionnaire of WoON 2009 are
only answered by this group of respondents.
A multinomial logistic regression model is used to analyse how various
personal and household characteristics and reasons for moving influence the type of
rural area people move to. The dependent variable in this model consists of the
categories ‘moved to less‐popular’, ‘moved to average’ and ‘moved to popular’ rural
areas (reference category).
Following Hjort and Malmberg (2006), the personal and household
characteristics we selected as independent variables are being part of a couple,
having children, age, income and education. In addition, we also selected having
moved from an urban area and having moved from within the northern
Netherlands. Because the analysis of migration patterns presented earlier showed
that popular rural areas attract more movers from urban areas and from outside the
North, we want to investigate whether this effect remains when other characteristics
are controlled for.
We also selected reasons for moving as independent variables, following
Stockdale (2006a), who showed that the motivations for in‐migration to peripheral,
depopulating rural areas differ from those for in‐migration to repopulating areas.
Unfortunately, the WoON questionnaire is designed in such a way that if people
have moved to their current address for personal reasons (marriage, living together,
divorce, leaving the parental home), they are not asked whether they also had other
motivations, work for example. Respondents who did not move for personal reasons
could choose more than one motive from the questionnaire. This may have resulted
in an underestimation of some of the reasons for moving in the data.
45
Table 5. Multinomial logistic regression analysis of having moved to less‐popular, average or
popular rural areas.
Dependent variable: Less‐popular
rural areas
Average rural
areas
Independent variables: B S.E. B S.E.
Couple (couple=1, single=0) 0.543 0.616 0.076 0.530
Children (yes=1, no=0) 0.467 0.496 0.134 0.446
Age ‐0.028 0.018 ‐0.007 0.015
Income ‐0.017 0.017 0.000 0.012
Higher education (yes=1, no=0) ‐1.235** 0.497 ‐0.868** 0.427
Moved from urban area (urban=1, rural=0) ‐1.541*** 0.472 ‐1.480*** 0.412
Moved from North of the Netherlands (yes=1, no=0) ‐0.859 0.584 ‐0.533 0.527
Moved for personal reasons (yes=1, no=0) 0.006 0.564 ‐0.735 0.539
Moved for work reasons (yes=1, no=0) 0.936 0.940 0.628 0.895
Moved because of previous dwelling (yes=1, no=0) ‐0.678 0.942 ‐0.614 0.801
Moved because of previous residential environment (yes=1,
no=0)
‐0.198 0.747 ‐0.453 0.633
Moved to live closer to family and friends (yes=1, no=0) 1.695* 0.884 0.089 0.930
Constant 2.434 1.216 1.709 1.088
N 170
‐2 log likelihood 315.086
2 49.770***
* p < 0,.10; ** p < 0,05; *** p < 0,.01
Source: WoON 2009
The analysis (see Table 5) shows that more highly‐educated people and people
moving from urban areas more often move to popular rural areas compared to less‐
popular and average rural areas. The other background characteristics do not
influence the probability of ending up in one of the three types of areas, while
income in particular could be expected to be a determining factor, given that the
areas are defined based on house prices. This outcome does seem to fit with the
‘classical’ counter‐urbanisation picture of middle‐class people moving from the city
to rural areas (Halfacree, 2008; Stockdale, 2006a). Apparently, popular rural areas
are more likely to attract these groups than less‐popular rural areas and average
rural areas.
46
However, because of the expected relationship between education and income
we also conducted the analysis without including higher education as an
independent variable. Then the results show that a lower income leads to an
increased probability to move to less‐popular areas compared to popular areas, but
the coefficient is only just significant (B = ‐0.032, p = 0.097). Apparently, the effect of
income is partly included in the effect of higher education. Yet, the effect of income
is small. It is not the important explanatory factor you would expect when
comparing areas with different levels of housing prices.
People who moved to their current address to live closer to family and friends
more often move to less‐popular areas compared to popular areas. This corresponds
with the findings of Stockdale (2006a) and Grimsrud (2011) that personal
motivations, such as a first home and marriage or family ties, were important
motives for migrating into peripheral, depopulating rural areas. Therefore, based on
motives, migrants who move to less‐popular rural areas in the Netherlands seem to
resemble migrants moving to peripheral rural areas.
The importance of the motive to live closer to family and friends could also
indicate that part of the migration flow into less‐popular areas consists of return
migration. Niedomysl and Amcoff (2010) show, in the context of internal return
migration in Sweden, that social relationships are an important motivation for
return. More specifically, proximity to family and friends is often mentioned by
return migrants as a motive. Unfortunately, this idea cannot be tested using the
WoON 2009 data. Except for the former place of residence, other former places of
residence of the respondents and their place of birth are unknown.
2.9 Conclusion
In the introduction to this paper we stated that migration patterns in rural areas
within core regions of Europe cannot be explained sufficiently by the prevailing
counter‐urbanisation models alone. We have tried to empirically underline this
statement by analyzing migration flows, characteristics and motives of people
moving to less‐popular, average and popular rural areas for living in the northern
47
Netherlands. Migration statistics show that migration numbers in the different types
of rural areas are largely comparable, though on the basis of house prices it can be
concluded that less‐popular areas are in general perceived as less attractive to live
in. This shows that amenities do not have the same meaning to all migrants and less‐
popular areas also offer something that attracts people.
Considering migration statistics, it appears that less‐popular areas attract a
larger share of movers from within the northern Netherlands, a larger influx from
rural areas and generally a younger group of movers. The analysis of survey data
showed that people who moved to their current address motivated by a desire to
live closer to family and friends have a higher probability of moving to less‐popular
areas. Popular areas attract a larger share of movers from outside the northern
Netherlands, a larger share from urban areas and in general a slightly older group of
migrants. Multivariate analysis showed that more highly‐educated people and
people moving from urban areas more often move to popular rural areas compared
to less‐‐popular and average rural areas.
The reason why popular areas attract more highly‐educated people could be
their proximity to the central part of the Netherlands. This provides easier access to
a more diverse labour market and the networks related to it.
An explanation of why popular rural areas receive more migrants from urban
areas can be found in the attractiveness of the semi‐open landscape and the presence
of nature in these areas. This can be related to representations or images of rural
areas that people have. Previous research has shown that in representations of urban
residents, visual aspects relating to landscape and other environmental
characteristics are more dominant (Steenbekkers et al., 2008; Haartsen, 2002), while
sociocultural values relating to the countryside way of life are more dominant in
representations of rural residents. The quality of the landscape is probably of lesser
importance to rural residents compared to the other characteristics of rural areas.
In addition to representations of ‘the rural’ in general, representations of
specific rural areas could also be important in explaining migration decisions. Such
representations could partly be based on the areas’ accessibility. From a regional
perspective, all rural areas in the northern Netherlands have similar access to the
48
northern cities. However, from the national perspective, the less‐popular areas are
more peripherally located. This fact, combined with a landscape that in general is
perceived as less attractive, could lead to less positive representations of less‐
popular areas. From earlier research, we know that representations are constructed
in direct and indirect contact with an area, and that familiarity with an area creates
more positive representations (Haartsen, 2002). The importance of the proximity of
family and friends for migrants to less‐popular rural areas indicates familiarity with
the area, and this could have resulted in more positive representations of the area.
To conclude, our study has shown that within core regions in Europe, the
classical counter‐urbanisation model does not explain all rural population change.
The characteristics of the movers to popular rural areas do fit very well within the
counter‐urbanisation story, while less‐popular rural areas share personal reasons as
an important motive for in‐migration with more remote rural areas in Europe.
Apparently, rural areas in core regions in Western Europe do vary in terms of
amenities and perceived remoteness. Our study indicates that representations of
what rural is, and concepts of periphery and remoteness have to be viewed within
the local, regional and national context. Research into rural population change in
both core regions and sparsely populated countries should consider these different
contexts, to be able to acknowledge the variety of ways amenities and peripherality
are perceived by different people.
For further research, we believe we can identify some other potentially relevant
aspects and questions. Firstly, we do not yet know enough about the motives of
migrants. The database we used mixes motives for moving with motives for location
choice, while the reasons for leaving somewhere can very well differ from the
motives for choosing a certain destination (see also Bolton and Chalkley, 1990). It
would also be interesting to know the birth place and residential history of migrants,
to investigate whether return migration and familiarity with rural areas and rural
living plays a role in migration to less‐popular areas. Finally, knowledge of the
representations people have of ‘the rural’ in general and different rural areas more
specifically, could help us to understand migration movements towards different
types of areas.
49
References
Andersen HS. 2009. Explanations for long‐distance counter‐urban migration into
fringe areas in Denmark. Population, Space and Place. DOI: 10.1002/psp.568.
Appleton J. 1975. The Experience of Landscape. Wiley: London.
Argent N, Smailes P, Griffin T. 2007. The amenity complex: towards a framework for
analysing and predicting the emergence of a multifunctional countryside in
Australia. Geographical Research 45: 217‐232.
Benson M, O’Reilly K. 2009. Migration and the search for a better way of life: a
critical exploration of lifestyle migration. The Sociological Review 57: 608‐625
Bollman RD, Briggs B. 1992. Rural and small town Canada: an overview. In Rural
and Small Town Canada, Bollman RD (ed.). Thompson Educational Publishing:
Toronto.
Bolton N, Chalkley B. 1990. The rural population turnround: a case‐study of North
Devon. Journal of Rural Studies 6: 29‐43.
Boyle P, Halfacree K. 1998. Migration into Rural Areas: Theories and Issues. John Wiley
& Sons: Chichester.
Brons LJ. 2005. Rethinking the Culture‐Economy Dialectic. Groningen University:
Groningen.
Brons LJ. 2006. Indirect measurement of regional culture in the Netherlands.
Tijdschrift voor Economische en Sociale Geografie 97: 547‐566.
Champion AG (ed.). 1989. Counterurbanization: the Changing Pace and Nature of
Population Deconcentration. Edward Arnold: London.
Costello L. 2007. Going bush: the implications of urban‐rural migration. Geographical
Research 45: 85‐94.
Elbersen B. 2001. Nature on the Doorstep: the Relationship between Protected Natural
Areas and Residential Activity in the European Countryside. Wageningen: Alterra.
Fitchen JM. 1995. Spatial redistribution of poverty through migration of poor people
to depressed rural communities. Rural Sociology 60: 181‐201.
Glaeser EL, Shapiro JM. 2003. Urban growth in the 1990s: is city living back? Journal
of Regional Science 43: 139‐165.
50
Gkartzios M, Scott M. 2009. Residential mobilities and house building in rural
Ireland: evidence from three case studies. Sociologia Ruralis 50: 64‐84.
Grimsrud GM. 2011. How well does the ‘counter‐urbanisation story’ travel to other
countries? The case of Norway. Population, Space and Place. DOI:
10.1002/psp.655.
Guimond L, Simard M. 2010. Gentrification and neo‐rural populations in the Québec
countryside: representations of various actors. Journal of Rural Studies xx: 1‐16.
DOI: 10.1016/j.jrurstud.2010.06002.
Haartsen T. 2002. Platteland: Boerenland, Natuurterrein of Beleidsveld? Een Onderzoek
naar Veranderingen in Functies, Eigendom en Representaties van het Nederlandse
platteland. Koninklijk Nederlands Aardrijkskundig
Genootschap/Rijksuniversiteit Groningen: Utrecht/Groningen.
Haartsen T, Huigen PPP, Groote P. 2003. Rural areas in the Netherlands. Tijdschrift
voor Economische en Sociale Geografie 94: 129‐136.
Haartsen T, Strijker D. 2010. Rural youth culture: keten in the Netherlands. Journal of
Rural Studies 26: 163‐172.
Haartsen T, Venhorst V. 2010. Planning for decline: anticipating on population
decline in the Netherlands. Tijdschrift voor Economische en Sociale Geografie 101:
218‐227.
Halfacree K. 1994. The importance of ‘the rural’ in the constitution of
counterurbanization: evidence from England in the 1980s. Sociologia Ruralis
34: 164‐189.
Halfacree K. 2008. To revitalise counterurbanization research? Recognising an
international and fuller picture. Population, Space and Place 14: 479‐495.
Halliday J, Coombes M. 1995. In search of counterurbanisation: some evidence from
Devon on the relationship between patterns of migration and motivation.
Journal of Rural Studies 11: 433‐446.
Heins S. 2002. Rurale Woonmilieus in Stad en Land: Plattelandsbeelden, Vraag naar en
Aanbod van Rurale Woonmilieus. Eburon: Delft.
Hjort S, Malmberg G. 2006. The attraction of the rural: characteristics of rural
migrants in Sweden. Scottish Geographical Journal 122: 55‐75.
51
Hugo G, Bell M. 1998. The hypothesis of welfare‐led migration to rural areas: the
Australian case. In Migration into rural areas, Boyle P, Halfacree K (eds.).
Wiley: Chichester; 107‐133.
Hugo GJ, Smailes PJ. 1985. Urban‐rural migration in Australia: a process view of the
turnaround. Journal of Rural Studies 1: 11‐30.
Ilbery B (ed.). 1998. The geography of rural change. Pearson Education Limited:
Harlow.
Johnson KM, Beale CL. 1994. The recent revival of widespread population growth in
nonmetropolitan areas of the United States. Rural Sociology 59: 655‐667.
Kasimis C, Papadopoulos AG, Zacopoulou E. 2003. Migrants in rural Greece.
Sociologia Ruralis 43: 167‐184.
Lindgren U. 2003. Who is the counter‐urban mover? Evidence from the Swedish
urban system. International Journal of Population Geography 9: 399‐418.
Luttik J. 2000. The value of trees, water and open space as reflected by house prices
in the Netherlands. Landscape and Urban Planning 48: 161‐167.
Marlet G. 2009. De Aantrekkelijke Stad. VOC Uitgevers: Nijmegen.
Marshall N, Burnley I, Murphy P, Hugo, G. 2004. Migration of Income Support
Recipients from Non‐metropolitan NSW and SA into Metropolitan Sydney and
Adelaide. The Australian Housing and Urban Research Institute: Melbourne.
McGranahan DA. 2008. Landscape influence on recent rural migration in the U.S.
Landscape and Urban Planning 85: 228‐240.
Mitchell CJA, Bunting TE, Piccioni M. 2004. Visual artists: counter‐urbanites in the
Canadian countryside? The Canadian Geographer 48: 152‐167.
Milbourne P. 2007. Re‐populating rural studies: migrations, movements and
mobilities. Journal of Rural Studies 23: 381‐386.
Niedomysl T, Amcoff, J. 2010. Why return migrants return: survey evidence on
motives for internal return migration in Sweden. Population, Space and Place.
DOI:10.1002/psp.
Philips M. 1998. Investigations of the British rural middle classes‐part 2:
fragmentation, identity, morality and contestation. Journal of Rural Studies 15:
317‐330.
52
Selwood J, Curry G, Jones R. 1996. From the turnaround to the backlash: tourism
and rural change in the Shire of Denmark, Western Australia. Urban Policy and
Research 14: 215‐225.
Selwood J, Curry G, Koczberski G. 1995. Structure and change in a local holiday
resort: Peaceful Bay, on the Southern coast of Western Australia. Urban Policy
and Research 13: 149‐157.
Shumway JM, Otterstrom SM. 2001. Spatial patterns of migration and income
change in the mountain west: the dominance of service‐based, amenity‐rich
counties. Professional Geographer 53: 492‐502.
Steenbekkers A, Simon C, Vermeij L, Spreeuwers W. 2008. Het Platteland van Alle
Nederlanders: Hoe Nederlanders het Platteland Zien en Gebruiken. Sociaal en
Cultureel Planbureau: Den Haag.
Stockdale A. 2006a. Migration: Pre‐requisite for rural economic regeneration? Journal
of Rural Studies 22: 354‐366.
Stockdale A. 2006b. The role of a ‘retirement transition’ in the repopulation of rural
areas. Population, Space and Place 12: 1‐13.
Stockdale A, Findlay A, Short D. 2000. Repopulation of rural Scotland: opportunity
and threat. Journal of Rural Studies 16: 243‐257.
Terluin I, Slangen LHG, Van Leeuwen, ES, Oskam AJ, Gaaff, A. 2005. De
Plattelandseconomie in Nederland: een Verkenning van Definities, Indicatoren,
Instituties en Beleid. LEI: Den Haag.
Ulrich RS. 1986. Human responses to vegetations and landscapes. Landscape and
Urban Planning 13: 29‐44.
Van Dam F, Heins S, Elbersen BS. 2002. Lay discourses of the rural and stated and
revealed preferences for rural living: some evidence of the existence of a rural
idyll in the Netherlands. Journal of Rural Studies 18: 461‐476.
Van den Berg AE, Koole SL, Van der Wulp NY. 1998. Environmental preference and
restoration: (how) are they related? Journal of Environmental Psychology 23: 135‐
146.
Visser P, Van Dam F. 2006. De Prijs van de Plek: Woonomgeving en Woningprijs. Nai
Uitgevers/Ruimtelijk Planbureau: Rotterdam/Den Haag.
53
Walmsley DJ, Epps WR, Duncan CJ. 1998. Migration to the New South Wales North
Coast 1986‐1991: Lifestyle motivated counterurbanisation. Geoforum 29: 105‐
118.
Westlund H. 2002. An unplanned green wave: settlement patterns in Sweden during
the 1990s. Environment and Planning A 34: 1395‐1410.
Woods M. 2005. Rural Geography. Sage Publications: London.
Yu K. 1995. Cultural variations in landscape preference: comparisons among
Chinese subgroups and Western design experts. Landscape and Urban Planning
32: 107‐126.
54
55
3. Migration to less‐popular rural areas in the Netherlands:
Exploring the motivations3
Abstract
Migration into rural areas is often explained in terms of the rural idyll, the attraction of the
countryside with its less hurried way of life in a quiet, spacious and green environment.
However, this migration phenomenon has mostly been researched in attractive, amenity‐
rich, popular rural areas. This paper investigates the characteristics and motivations of
migrants to less‐popular rural areas using survey data (N=664) for four municipalities in the
North of the Netherlands. Our study shows a young group of in‐migrants with relatively
low incomes, but also a large proportion of working people and a considerable number of
highly educated movers. Separating the motivations for choosing to live in a rural area in
general from the motivations for choosing this specific rural area reveals that while the pull of
the rural idyll is an important motivation for moving to a rural area in general, the reasons
for choosing the specific rural area are a mixture of housing characteristics, the physical
qualities of the environment, personal reasons and the low house prices in the area.
Combining the motivations with the characteristics of the movers reveals the diversity
within the movers group. Our analysis shows a group of movers motivated to live close to
family and friends, consisting of return migrants, singles, the youngest and oldest age
groups and also the lowest income group. The physical qualities of the environment attract
a group of highly educated movers, people with high incomes and people aged between 35
and 64. The motivation of housing characteristics, referring in most cases to the availability
of a specific house, is mentioned by a wide range of movers, but in particular by people
moving from urban areas.
3.1 Introduction
Residential use has become an emerging function of the post‐productivist or
consumption countryside. Processes of suburbanisation and counterurbanisation
have brought new inhabitants to the rural. These migration flows to, often attractive,
3 This chapter is reprinted from: Bijker RA, Haartsen T, Strijker, D. 2012. Migration to less‐popular
rural areas in the Netherlands: Exploring the motivations. Journal of Rural Studies 28: 490‐498.
56
rural areas are predominantly viewed as movements of middle‐class people from
urban areas in search of a rural lifestyle (Milbourne, 2007). However, it has
increasingly been acknowledged that the spatial context ‘needs to be conceptualised
as contingent’ in migration research (Findlay, 2005: p. 429). Simultaneously,
migration has become more diverse and complex due to developments in, for
instance, mobility and IT (Smith and King, 2012). Rural scholars recently started
questioning the ‘classical’ view of counterurbanisation as the only conceptualisation
of rural migration, as it does not acknowledge that rural areas can differ in their
popularity for living, that rural areas can also attract other types of migrants with
different motives, and that mobilities have become much more diversified (e.g.
Bijker and Haartsen, in press; Grimsrud, 2011; Halfacree, 2008, 2012; Milbourne,
2007; Woods, 2011). Empirical evidence is needed ‘for shedding light on the different
scales, rates and outcomes of diverse processes of migration’ (Smith and King, 2012:
p. 127) and for looking beneath the ‘meta‐narratives of rural population change’
(Milbourne, 2007: p. 382).
This paper aims to contribute to the debate on new migration theories by
providing new empirically founded insights into migration flows to the relatively
under‐researched less‐ popular rural areas. The limited existing studies of migration
to less‐popular rural areas indicate that such areas also attract movers other than the
middle‐class urbanites associated with the ‘counterurbanisation story’, ranging from
low‐income groups (Fitchen, 1995; Foulkes and Newbold, 2008) to a more mixed
group of movers (Andersen, 2011). However, the extent to which the typical
‘quality‐of‐life’ motivations connected to the counterurbanisation concept also hold
true for these migrants is not clear (Foulkes and Newbold, 2008; Stockdale, 2006). It
is also not known which motives apply to which type of migrants, and whether they
differ for short or long distance migrants.
In this paper we focus on less‐popular rural areas in the northern Netherlands.
They are specifically contextualised in a densely populated and urbanised country.
We are aware of the fact that the definition of less‐popular areas may differ for
different (national) contexts in terms of population density, physical characteristics
and distance to urban areas. Nonetheless, our research area shares characteristics
57
with less‐popular rural areas in other studies, such as a relatively peripheral location
and relatively low house prices (see e.g. Andersen, 2011; Fitchen, 1995; Foulkes and
Newbold, 2008; Stockdale, 2006). Moreover, based on previous research, we
concluded that representations of what is rural and notions of peripherality have to
be viewed within their specific national context (Bijker and Haartsen, in press).
In this study we investigate the characteristics and motives of migrants to less‐
popular rural areas in the Netherlands. We want to shed more light on the
motivations for moving to less‐popular areas by disentangling them into reasons for
leaving the previous place of residence, motivations for moving to the rural in
general and motivations for moving to the specific less‐popular rural area. Previous
research in a repopulating rural area has shown the usefulness of this approach
(Bolton and Chalkley, 1990). In addition, we take into account the
multidimensionality of motives for moving, which makes it easier to reveal the
importance of the rural character of the destination (Halfacree, 1994; Halliday and
Coombes, 1995). By combining the movers’ characteristics with their motivations,
we explore which motivations are mentioned by whom.
We start by providing a literature overview of the characteristics and motives
of rural migrants. The methodology section describes our research area and the
survey design. The sections on the empirical results present the people who move to
less‐popular rural areas in the Netherlands, why they move and which groups of
movers mention which motivations. In the discussion section, we try to relate our
empirical findings to the wider debate on rural migration.
3.2 Migration into rural areas: who?
Rural in‐migration in general is often associated with the movement of middle‐class
families from the city who are either retired or commute to nearby urban centres for
employment (Smith and Phillips, 2001; Stockdale, 2006). In the Netherlands
migration to rural areas complies to a large extent with this ‘counterurbanisation
story’. Around half of the movers to rural areas moved from urban areas, middle
58
and high‐income groups are overrepresented, and around 40% of the movers are
highly educated (Steenbekkers et al., 2006; Steenbekkers et al., 2008).
Nevertheless, in an international context, several authors have made clear that
migration to rural areas in general is more diverse than this classic image of
counterurbanisation. Firstly, in addition to urban‐to‐rural migration, there is rural‐
to‐rural migration (Gkartzios and Scott, 2009), and instead of consisting only of
longer distance movements, short distance relocations form a considerable share of
rural in‐migration (Walford, 2007). Secondly, it is not just the middle class moving to
the rural. Halfacree (2008) and Hoggart (2007) note the presence of quite a number
of working class counterurbanisers in rural England. Thirdly, not all urban‐to‐rural
migrants continue to work in cities. Findlay et al. (2000) challenge this − in their
words ‘one of the long‐held myths’ − by showing in their study that almost half of
all migrant heads of household in rural Scotland worked locally. Additionally, a
substantial share was not in active employment, leaving just a small group of long‐
distance commuters. Fourthly, there is a popular idea that it is mainly older age
groups who migrate to the countryside (Steenbekkers et al., 2006; Van Dam et al.,
2002). However, when examining the actual movers to rural areas in the
Netherlands, it appears that the younger age groups are in the majority, with only
16% being 65 or older. When they move, older people are relatively more likely to
move to rural areas, but they only represent a small proportion of the overall flow of
movers to rural areas (Steenbekkers et al., 2008).
When focusing on less‐popular rural areas, we see a picture that is even more
different from the ‘classic’ image described above. With regard to the origin and age
of the movers, Stockdale (2006) found that most in‐migrants in depopulating,
peripheral areas in rural Scotland moved short distances, and that there was a mix of
moves from cities, from smaller settlements and between similar sized rural
settlements. In her study most respondents (75%) had moved before their fiftieth
birthdays. This corresponds with the findings in a previous study in the northern
Netherlands based on migration statistics and limited secondary data, which found
that less‐popular areas attract more migrants from within the northern Netherlands,
from rural areas and also a younger group of movers (Bijker and Haartsen, in press).
59
Furthermore, less‐popular areas do not seem to attract the middle class. In
Stockdale’s study, skilled manual, trade‐type occupations dominated, with few
being employed in the professional or managerial sectors. In the previously
mentioned study in the northern Netherlands, less‐popular rural areas attracted
more migrants with lower levels of education. Furthermore, impoverished rural
areas in the U.S., characterised by ample availability of cheap rental housing, attract
large groups of migrants with low incomes, varying from working poor (Foulkes
and Newbold, 2008) to welfare benefits recipients (Fitchen, 1995). Nonetheless, a
more diverse group of in‐migrants were found moving into fringe areas in
Denmark, i.e. areas experiencing population decline and a low level of economic
activity (Andersen, 2011). In addition to some groups with quite low incomes that
Andersen assumes migrate for lower housing costs, these areas also attract movers
with middle or high incomes, partly consisting of couples with children moving to a
detached house, something they are probably unable to afford in Denmark’s growth
areas.
3.3 Migration into rural areas: why?
The motives for migration to rural areas in general can be subdivided into three
groups: residential, household, and work and education motives (Van Dam et al.,
2002). The first group, residential motives, refers to the quality of the housing and
residential environment. The quality of the residential environment is often seen as
the most important pull factor of rural areas (Deller et al., 2001; Gkartzios and Scott,
2009; Halfacree, 1994; Halliday and Coombes, 1995; Walmsley et al., 1998).
Following Halfacree (1994), physical and social features of the residential
environment can be identified. Physical features are for example space, fresh air and
an attractive landscape, while social features include friendly people, peaceful living
and less crime. Movers to rural areas in the Netherlands most often mention the
typical rural idyllic arguments such as space and quietness, after housing
characteristics, as reasons for choosing their place of residence (Steenbekkers et al.,
2008).
60
In addition to the residential environment, housing characteristics can also be
an important reason for moving to the countryside, for example a better or bigger
house, or a large garden (Gkartzios and Scott, 2009). For movers to rural areas in the
Netherlands, the availability of a specific house or certain type of house was the
most important reason for choosing the current place of residence (Steenbekkers et
al., 2008). Housing‐related reasons are often associated with short‐distance moves
(Halfacree, 1994; Mulder, 1993).
The second group of motives, the household motives, refers to moves
motivated by life events such as moving in together, getting married, having
children or getting divorced (Van Dam, 2000). The desire to live closer to family as a
reason for moving also belongs to this category. Proximity to family and friends is
mentioned by 11% of movers to rural areas in the Netherlands as having been
important to their choice of current residence (Steenbekkers et al., 2008).
The third group of motives is related to employment and education. Migration
motivated by a change of job, the desire to live closer to work, or to get an education,
fits into this category (Van Dam, 2000). In particular, work motives play an
important role in rural migration. Employment‐related reasons are often associated
with long‐distance moves (Halfacree, 1994). In the Netherlands, migrants originating
from rural areas were more often motivated by considerations related to their work
than movers from urban areas (Van Dam et al., 2002).
A special motive for rural migration is return migration. Different percentages
of return migrants have been found in rural migration studies, ranging from only
9% in North Devon (Bolton and Chalkley, 1990) to as much as 50% in amenity‐rich
rural areas in the Netherlands (Elbersen, 2001). In these cases return migration refers
to moving back to a place where a person lived previously. Niedomysl and Amcoff
(2011) found that return migrants are more often motivated by social reasons
compared to non‐return migrants. It is particularly the proximity to family and
friends that motivates them to return. In addition to return migration to a specific
place, more general ‘return to the rural’ types of in‐migration have also been found.
For example, in the three Irish rural case study areas of Gkartzios and Scott (2009), a
large majority of the migrants appeared to have grown up in rural areas. In the
61
Netherlands Feijten et al. (2008) have also shown that people who had previously
lived in rural areas have a significantly higher probability of moving to rural areas
compared to people without experience of rural life.
When focusing specifically on motives for migrating to less‐popular rural
areas, it appears that residential, household, and work and education motivations
for moving also play a role in these areas, but the relative importance of the different
motivations seems to differ. While in general the quality of the residential
environment is an important pull factor for in‐migration to rural areas, household
motivations such as marriage, divorce, setting up home for the first time and
employment considerations appear to be the main motivations for moving to rural
depopulating areas in Scotland, while housing was another important motivation
(Stockdale, 2006). In their qualitative case studies into depressed rural communities
attracting low‐income groups, Fitchen (1995) and Foulkes and Newbold (2008) also
show the importance of housing reasons, mainly relating to the availability of
affordable housing. In these studies, household motivations were also important: a
considerable share of migrants was attracted by the proximity to family and friends.
However, with regard to the importance of the pull of the countryside as a way
of life for in‐migration to less‐popular rural areas, the studies show inconsistent
results. While Stockdale (2006) found quality‐of‐life‐considerations to be of little
importance, accounting for no more than 10% of moves, the in‐depth interviews
conducted by Foulkes and Newbold (2008) reveal that in addition to the attraction of
low housing costs, the rural character of the community did play a role in the
decision to move there. Moving to these communities represented an opportunity to
satisfy the desire to live in a rural setting. Many migrants viewed their move to the
area as a step closer to their ultimate residential goal of living out in the country on a
large piece of property (Foulkes and Newbold, 2008).
These different results could be caused by the different methodological
approaches of the studies. Motives for moving are never one‐dimensional and
previous research has shown that when asking for the single most important motive
for moving – as Stockdale does in her study – underlying motivations and
62
preferences can remain invisible (Van Dam, 2000), resulting in an underestimation of
the importance of the rural character of the destination (Halfacree, 1994; Halliday
and Coombes, 1995). Foulkes’ and Newbold’s (2008) qualitative method is probably
better able to reveal this significance of the rural environment.
Another methodological issue concerning the analysis of migration motives is
the distinction between motives for leaving a specific location from motives for
moving to a specific location. Bolton and Chalkley (1990) showed the importance of
this distinction when concluding that the reasons for leaving the previous area of
residence tended to relate to lifestyle, personal or environmental factors, while the
reasons for choosing North Devon were more often about jobs and house prices. In
Stockdale’s analysis, motives for leaving the previous place of residence are mixed
with motives for moving to the specific area, probably resulting in an
underestimation of the importance of the quality of the residential environment. We
think that addressing these two methodological issues using a quantitative approach
could help clarify the motivations of movers to less‐popular areas.
3.4 Methodology
The data we present in this paper were collected in less‐popular rural areas in the
northern Netherlands. Even though in comparison with other European countries,
the Netherlands is a densely populated and urbanised area, a considerable part of
the country is defined as rural by the Dutch population (Haartsen et al., 2003). In
this study we focus on the northern Netherlands because, based on address density
and the perception of the Dutch people, it is the most rural part of the country. The
popularity for living of our rural areas is defined using average house prices per
municipality. House prices are an indication of what people are willing to pay for
houses and their surroundings (Luttik, 2000; Visser and Van Dam, 2006), thus
reflecting the value buyers attach to them. Therefore, we consider the average house
prices in an area as an indicator of the popularity of an area for living. This average
house price incorporates the residential environment as well as housing
characteristics such as size and quality. The use of house prices instead of in‐
63
migration numbers, which are often used in studies of rural areas’ popularity
(Argent et al., 2007; McGranahan, 2008), is more appropriate to the Dutch context. In
the Netherlands population growth at the municipality level is strongly influenced
by government development policy (Marlet, 2009). New housing is only allowed in
a limited number of areas, while in others, building houses is very restricted,
especially in rural areas, both in the open countryside and in villages (Van Dam et
al., 2002). These restrictions make it impossible for the Dutch housing market to
react to changes in demand for new houses, which leads to price increases rather
than increased in‐migration when there is increasing demand (Marlet, 2009). Rental
prices differ little between regions in the Netherlands, which is why these are not
considered in defining the popularity of rural areas for this study.
Based on house prices, we distinguish three types of rural areas in the northern
Netherlands: less‐popular rural areas, average rural areas and popular rural areas.
These areas are considered rural based on Statistics Netherlands’s national standard
for the degree of urbanisation, the so‐called address density, consisting of
municipalities with an average of less than 1,000 addresses per square kilometre.
The average house prices per municipality for the rural municipalities in the
northern Netherlands varied from EUR 142,600 to EUR 340,000 in 2008. We
performed a cluster analysis on this average house price per municipality resulting
in three clusters, which we called less‐popular rural areas (average price EUR
174,828), average rural areas (EUR 212,731) and popular rural areas (EUR 273,170)
(see also Bijker and Haartsen, in press). When looking at average house prices per
municipality in the whole of the Netherlands in 2008, the less‐popular areas, which
are the focus of this paper, are also the lowest in ranking. Since 2008 the crisis in the
housing market has resulted in price drops and decreasing numbers of transactions
in most municipalities in the Netherlands, including those in the northern
Netherlands. While it is a process that still is taking place, so far this has not
changed the position of the less‐popular areas with regard to the whole of the
Netherlands and the general pattern of popularity within the North has not changed
substantially either.
64
Figure 1 shows that the areas with the lowest house prices, the less‐popular
areas, are located along the northern and eastern borders of the northern
Netherlands. They can be characterised as having an open, marine clay landscape
with a large proportion of large‐scale arable farmland. Compared to the popular
areas, they have relatively fewer natural and recreational areas and they have a
relatively low proportion of hotel and catering industry employment. They are
relatively distant from the centre of the Netherlands and the people who inhabit
these areas have a lower average household income than those in the average and
popular rural areas. In these less‐popular areas, both population decline and a
decline in the number of households are expected (Bijker and Haartsen, in press)
and in some municipalities population decline is already occurring (Haartsen and
Venhorst, 2010).
Figure 1. Three types of rural areas in the northern Netherlands
From the less‐popular rural areas, we selected four municipalities as study
areas: De Marne (10,587 inhabitants in 2009), Menterwolde (12,514) and Reiderland
65
(6,988) in the province of Groningen and Ferwerderadiel (8,825) in the province of
Friesland (see Figure 1). In the period 2005‐2009, all these municipalities experienced
population decline. None of the four municipalities has a large town within its
boundaries and they do not border one of the larger cities in the North. Therefore, in
the Dutch context all migrants to these municipalities can be considered to be rural
migrants rather than movers to a larger settlement or suburbanisers. The selected
municipalities are roughly equally distant in terms of commuting distance to the
larger cities in the North (around 40 minutes). While the landscape can be
characterised as open in all four municipalities, due to different soil types (marine
clay and peat) and historical background, the municipalities represent different
landscape types in the North.
In cooperation with the municipal administrations, we selected all households
who had moved into the municipality from another municipality in the period 2005‐
2009 based on the municipal population registers. In this selection owner‐occupiers
as well as renters were included. The relatively short period of 5 years was chosen to
reduce the risk of memory‐recall problems of the respondents and in particular post
hoc rationalisation (see e.g. Walmsley et al., 1998).
In autumn 2009 a postal questionnaire was sent to all the households that met
the selection criteria. In an accompanying letter we asked whether a household
member aged over 18 could complete the questionnaire. A total of 664
questionnaires were returned, representing a response rate of 20%. Statistics
Netherlands provided data on the age distribution of the entire group of movers to
these municipalities. In comparison to these data, it appeared that in our sample the
older age groups were over‐represented, while the youngest age group was under‐
represented. To obtain a representative age distribution we weighted the sample
with regard to age for the descriptive analysis.
The questionnaire covered a variety of topics: residential history, demographic
and socioeconomic characteristics, housing characteristics, previous experience with
the area and the motivations that were important in the migration decision.
Following Van Dam (2000) and Bolton and Chalkley (1990), we separated
motivations for leaving the previous residence from the motives for choosing the
66
destination area. Regarding the motives for choosing the destination area, we
distinguished between choosing to live in a rural area in general and choosing to
live in the specific less‐popular rural area. Both questions on the choice of the
destination area were open‐ended, to take into account the multidimensionality of
motives for moving. Open‐ended questions offer the possibility of receiving
spontaneous responses and avoid the bias that can be caused by suggesting
responses, thus enabling a more open‐minded and less biased approach to studying
migration motives (Halfacree, 2004; Niedomysl and Malmberg, 2009). Moreover,
Niedomysl and Malmberg (2009) have shown that coder variability when coding
open‐ended questions on migration motives is relatively low.
3.5 Characteristics of migrants to less‐popular rural areas
The demographic characteristics of migrants to the less‐popular rural areas in our
survey are summarised in Table 1. It shows that the four selected municipalities in
the northern Netherlands mainly attract a young group of movers. Almost half of
the migrants are aged between 20 and 34, while only 6% are over the retirement age
of 65. In addition, compared to migrants to rural areas in general in the Netherlands
(Steenbekkers et al., 2008), the less‐popular areas seem to attract a younger group of
movers. This probably also explains the fact that the share of couples without
children is relatively high (47%). Covering 43% of our respondents, the proportion
of movers with higher education is comparable to movers to rural areas in the
Netherlands in general. Movers to rural areas seem to be quite highly educated,
considering the fact that in the Netherlands as a whole, only around a third of the
population has completed higher education (Steenbekkers et al., 2008).
67
Table 1. Characteristics of in‐migrants to the study areas (%) (n=664)
Age % %
20‐34 48 Lived in municipality before 16
35‐44 20
45‐54 13 Lived in a rural area before 88
55‐64 13
65 < 6 Moved from urban area 43
Household composition
Couple without children 47 Previous residence
Couple/single parent with
children
37 Surrounding municipalities 28
Single person household 16 Elsewhere in the northern Netherlands 38
Level of education Elsewhere in the Netherlands 29
Lower or middle level of
education
57 Abroad 5
Higher education 43
From the analysis of the residential history of the respondents, it appears that
16% had lived in the municipality previously and can therefore be considered return
migrants (see Table 1). This percentage does not seem high, given that Niedomysl
and Amcoff (2011) describe that in various Western countries, around a quarter of
all internal migrants can be categorised as return migrants. The more general ‘return
to the rural’ migration appears to be of much more significance to our respondents.
As many as 88% of the migrants in our study had lived in a rural area previously.
Around two thirds of these migrants had lived in the countryside more than half
their lives.
People may be already familiar with an area for various reasons. Not presented
in Table 1 but covered in the survey is the question of whether our respondents
already knew the area they moved to from personal experience. This was the case
for 61%, such as through living in the municipality or the surrounding
municipalities (25%), visiting friends and relatives (28%), visits for recreational
activities or holidays (9%), and through work‐related visits (9%). Considering the
previous places of residence (see Table 1), a relatively small percentage (43%) of our
respondents had moved from an urban area. For rural areas in the Netherlands as a
whole, this percentage is 57% (Steenbekkers et al., 2008). With regard to the location
68
of the previous residence, Table 1 shows a variety of movements, ranging from a
move over a longer distance from elsewhere in the Netherlands to a local move from
surrounding municipalities. The largest proportion (38%) moved from within the
northern Netherlands, which could be described as a regional move.
Table 2. Socioeconomic characteristics of in‐migrants to the study areas (%) (n=664)
Monthly household
income (net)
% Location of work1 % Economic sector1 %
< EUR 1250 12 In the municipality 15 Farming and fishing 5
EUR 1250 ‐ <2500 45 In surrounding
municipalities
15 Manufacturing 9
EUR 2500 ‐ <4000 34 In nearest city 30 Construction 6
≥ EUR 4000 9 Elsewhere in the province 15 Trade 7
Source of income Elsewhere in the North 10 Hotel and catering industry 2
Paid employment 64 Elsewhere in the
Netherlands
5 Transport and communication 5
Self‐employed 10 Several locations 3 Financial, business/other
services 19
Benefit 8 Other 8 Government 6
Pension 10 Education 9
Income partner 2 Health and social care 25
Student 2 Other 6
Other 4 1 Percentages do not sum to 100% due to rounding
Table 2 presents the socioeconomic characteristics of our respondents. Almost
half of the migrants in our sample earn between EUR 1250 and EUR 2500 per month.
A comparison with movers to rural areas in the Netherlands in general is impeded
by the different categorisation of incomes in Steenbekkers et al. (2008). However, it
remains clear that movers to less‐popular areas have lower incomes than movers to
the Dutch countryside have on average. A relatively large proportion (64%) is in
paid employment. The number of self‐employed (10%) and those on benefits (8%) is
comparable to movers to rural areas in the Netherlands in general. The movers work
in a variety of sectors. ‘Health and social care’ and ‘financial, business and other
services’ are the main sectors the migrants are employed in. Although the
proportion of people working in health and social care is high compared to the
69
Netherlands as a whole (based on Statistics Netherlands data), in the context of the
northern Netherlands this percentage is less surprising. The health and social care
sector is the largest employer in the northern Netherlands and employment in this
sector has been increasing in recent years (Schudde et al., 2010). Around a third of
the employed and self‐employed work in the municipality they live in or in
surrounding municipalities, which confirms that rural migration is indeed not
always connected to long‐distance commuting (Findlay et al., 2000).
The housing characteristics in Table 3 show that migrants to less‐popular rural
areas in the northern Netherlands predominantly moved to owner‐occupied
housing (77%). This is a larger proportion than the owner‐occupied housing
segment in the housing stock in the selected municipalities (65%, according to data
from Statistics Netherlands). It is also a higher proportion than generally observed
for movers to rural areas in the Netherlands as a whole (65%, Steenbekkers et al.,
2008). In terms of the type of house the migrants moved to, the majority (58%)
moved to a detached house, while 42% of the people living in rural areas in the
Netherlands live in a detached house (Steenbekkers et al., 2008). This could indicate
that the less‐popular areas studied here offer the possibility to satisfy certain
residential preferences for people who cannot afford this elsewhere, such as buying
a house in a rural setting or owning a detached house.
Table 3. Housing characteristics of migrants to the study areas (%) (n=664)
Type of tenure
Owner‐occupied 77
Rented house 23
Type of house
Terraced 21
Detached 58
Semi‐detached 18
Apartment 3
To summarise, our migrants seem to be more diverse than the low‐income
groups found to move to impoverished rural areas in the U.S. (Fitchen, 1995;
Foulkes and Newbold, 2008) and the movers to depopulating areas in Scotland
70
(Stockdale, 2006). Although they have a relatively low income in the Dutch context,
the lowest‐income group is only small. Combined with the relatively large
proportion of working people, the variety of sectors they work in, and the large
group with higher education qualifications, our movers seem to correspond better to
the more diverse group of movers to fringe areas in Denmark, which includes quite
a large proportion of migrants on low incomes, but also encompasses movers with
middle or high incomes trying to realise housing preferences they could not afford
in more expensive areas (Andersen, 2011). Based on the relatively high proportion of
owner‐occupied and detached houses, it appears that also in the less‐popular areas
we studied, housing preferences which are difficult to afford elsewhere can be
realised.
3.6 Motivations of migrants
Our questionnaire asked the migrants in the sample to answer three questions
related to their motivations for their move. The first was a closed question about
their reason for the move. Table 4 shows the reasons reported for moving. Because
respondents were allowed to give more than one answer, the percentages do not
add up to 100%. The move to less‐popular areas is most often triggered by changes
in the household career, with marriage or cohabitation as the main reason for
moving (33%). Nonetheless, residential reasons are also important. One‐fifth of the
migrants mentioned wanting to leave the city. Dissatisfaction with the previous
house or neighbourhood was also mentioned quite often. Finally, employment‐
related reasons, such as changes in the work situation and retirement, were also
mentioned. Remarkably, the ‘housing costs too high’ reason, which would have
been expected to be important due to the relatively low house prices in the area, was
not often mentioned. The answers in the ‘other’ category show that dissatisfaction
with the current dwelling or location is not always needed as a trigger for moving.
As part of the ‘other’ category, around 5% of the movers mentioned residential pull
factors – such as ‘realising our residential dream’, ‘building our own house’ and
‘living in the open’ – rather than push factors as their reason for moving. A small
71
number of movers did not even have plans to move, but just accidentally ‘fell in love
with the house’.
Table 4. Reasons for moving (%) (n=664)
Marriage/Cohabitation 33 Divorce/Decease of partner 5
Getting out of the city 22 Housing costs too high 5
Previous house 18 Living closer to family 3
Change in work situation 14 Children leaving home 2
Previous neighbourhood 12 Addition to the family 2
Setting up home for the first
time
8 Health 2
Retirement 7 Other 13
The answers to both open‐ended questions on motives for the choice of the
destination area were categorised on the basis of previous studies of rural migration
(e.g. Bolton and Chalkley, 1990; Gkartzios and Scott, 2009; Halfacree, 1994; Halliday
and Coombes, 1995; Walmsley et al., 1998) and also on what emerged from the
survey results. Table 5 shows that quietness is the main motivation for moving to a
rural area in general, directly followed by the physical qualities of the environment –
for example ‘space’, ‘nature’ and ‘fresh air’. The social qualities of the environment
are also important, such as ‘freedom’, ‘kind‐heartedness’, ‘friendliness’, ‘small‐scale’
and ‘social contacts in the village’. These motivations fit nicely with the elements
usually associated with the rural idyll, as do escaping the negative aspects of city life
included in the ‘city push factors’ motivation (for example ‘fed up with traffic jams
and crowdedness’ and ‘Rotterdam became too hectic and noisy’). Housing
characteristics are also important, referring for example to a ‘detached house’ or a
‘large garden’. Only one in ten of the respondents were motivated by lower house
prices. On the whole, it can be concluded that other residential motives dominate
the choice for rural living.
72
Table 5. Motivations for moving to a rural area (%) (n=664)
Quietness 43 Way of life 9
Physical qualities of the
environment
42 Work 8
Social qualities of the environment 17 Proximity of family of friends 7
Housing characteristics 13 Moving in with partner 6
Familiarity with the rural 12 For the children 4
City push factors 12 Other motivations 8
Low house prices 10
The respondents’ motivations for moving to this specific rural area are presented
in Table 6. Housing characteristics – and in most cases the availability of a specific
house – were the main motivation for moving to this specific rural area, with
answers such as ‘this house appealed to us’, ‘because of the house’ and ‘we found
the right house’. Here too, the physical qualities of the environment were important,
including ‘panoramic views’, ‘beautiful surroundings’, ‘space’ and ‘nature’. Sixteen
percent of the respondents had moved in with a partner, while about the same
proportion referred to another personal motivation: living close to family and
friends. Fourteen percent referred explicitly to the low house prices in the area as a
motivation for moving there. Work‐related reasons (‘closer to work’, ‘accessibility of
my work’) and familiarity with the area were also important. Some people stated in
a more general sense ‘I was familiar with the area’, while others specifically referred
to being a return migrant: ‘I grew up here’, ‘feeling at home’ and ‘wanted to return’.
Table 6. Motivations for moving to this specific rural area (%) (n=664)
Housing characteristics 24 Familiarity with the area 8
Physical qualities of the environment 21 Quietness 8
Moving in with partner 16 Location 8
Proximity to family and friends 15 Social qualities of the
environment
8
Low house prices 14 Other reasons 16
Work 12
In summary, while moving to the rural in general was mainly motivated by
residential reasons, moving to the specific rural area was motivated by a mixture of
73
residential, household and work‐related reasons. These motivations do not differ
that much from the reasons mentioned by movers to rural areas in the Netherlands
in general. They also referred to housing characteristics, space and green, work and
the proximity to family and friends as important reasons for choosing their place of
residence (Steenbekkers et al., 2008).
Compared to the movers to amenity‐rich rural areas in the Netherlands (Van
Dam et al., 2002), it is clear that the physical characteristics of the environment and
work‐related reasons are less important motives for moving to less‐popular rural
areas, while housing characteristics, affordable housing and personal reasons are
more important. This seems to support the conclusion of Stockdale (2006) that less‐
popular areas do indeed attract movers with different motives than popular rural
areas. However, migrants to less‐popular areas in the northern Netherlands do also
mention the physical characteristics of the environment, and their motivations for
moving to the rural in general refer to aspects of the rural idyll. Combined with the
motivation of the low house prices, this appears to correspond with the findings of
Foulkes and Newbold (2008) that these areas also offer the opportunity to fulfil the
dream of rural living for a group with relatively low incomes.
3.7 Linking characteristics and motivations
Following this overview of the background characteristics and motivations of the
movers to less‐popular rural areas, the two aspects of migration need to be linked –
which motivations are mentioned by whom? As noted previously, people can have
more than one motivation for moving to this specific rural area. We used logistic
regression analysis to characterise the people who mentioned a given motivation as
one of their motivations for choosing this area.
We estimated five logistic regression models, focusing on five of the six most
frequently mentioned motivations. The sixth motive, moving in with a partner, was
less interesting to analyse in the present context because it is more closely related to
the respondents’ life stage than to the area moved to.
74
The results of the logistic regression analyses are shown in Table 7. The
regression model estimates how various personal characteristics influence having a
particular motivation to move to this specific rural area. A positive B coefficient
means that an increased value on the independent variable increases the probability
of having this motivation. If the sign of the B coefficient is negative, an increase in
the value of the independent variable leads to a lower probability of having that
motivation.
The physical qualities of the environment are mentioned most often by highly
educated movers and people with a monthly household income of more than EUR
4000 per month compared with the lowest income group. People aged between 35
and 64 also mention this motivation more often than the youngest age group. People
who have lived in the municipality before are less often motivated by the physical
qualities of the environment. This corresponds with the finding of Niedomysl and
Amcoff (2011) that return migrants are less likely to choose their destination for
reasons related to the living environment.
Moving to the area because of housing characteristics or the availability of a
specific house is more often mentioned by people moving from an urban area.
People who had lived in the municipality previously mentioned this motivation less
often. The other background characteristics do not have a significant effect, meaning
that a diverse group of movers is attracted to the area by housing characteristics.
Based on the literature, housing reasons would be expected to be related to short‐
distance moves (Halfacree, 1994).
75
Table 7. Logistic regression analysis of motivations for moving to the specific rural area
Model 1 Model 2 Model 3 Model 4 Model 5
Dependent variable: Physical
qualities
Housing
characteristics
Proximity to
family and
friends
Low house
prices
Work
Independent variables: B B B B B
Moved from urban
area (urban = 1, rural =
0)
‐0.004 0.391* ‐0.150 ‐0.173 ‐0.472
Lived in municipality
before (yes = 1, no = 0)
‐0.831** ‐0.910** 0.694** ‐1.763** ‐0.155
Income
(ref. < EUR 1250)
EUR 1250‐<2500 0.307 ‐0.239 ‐0.618 0.778 0.108
EUR 2500‐<4000 0.122 0.271 ‐0.565 0.287 0.361
≥ EUR 4000 0.879* 0.679 ‐2.030* 0.608 0.020
Household
composition
(ref. couple without
children)
Couple/single
parent with
children
‐0.031 0.016 0.236 0.374 ‐0.086
Single household 0.020 ‐0.208 0.795** 0.654* ‐0.352
Previous residence (ref.
surrounding
municipalities)
Elsewhere in the
northern Netherlands
‐0.439 0.257 ‐0.534 0.364 ‐0.748**
Elsewhere in the
Netherlands
‐0.001 ‐0.211 0.220 0.870** ‐0.684*
Abroad ‐0.808 0.433 0.512 0.789 ‐0.483
Higher education (yes
= 1, no = 0)
0.453** 0.206 0.255 0.400 0.496*
Age (ref. 20‐34)
35‐44 0.803** ‐0.196 ‐0.854* ‐0.125 0.740**
45‐54 1.031*** 0.214 ‐1.100** ‐0.371 0.267
55‐64 1.351*** 0.032 ‐0.492 ‐0.786* ‐0.709
> 65 0.236 ‐0.097 0.081 ‐0.541 ‐1.860*
Constant ‐2.098 ‐1.342 ‐1.423 ‐2.835 ‐1.692
N 539 539 539 539 539
�2 53.699*** 29.416** 39.019*** 31.001*** 33.793***
Degrees of freedom 15 15 15 15 15
Nagelkerke R2 0.141 0.078 0.135 0.102 0.117
* p < 0.10; ** p< 0.05; *** p < 0.01
76
Moving to this specific rural area to live close to family and friends is
mentioned more often by people who had lived in the municipality previously. This
is the only motivation for which being a return migrant has a positive significant
effect. It seems that return migration to less‐popular rural areas is mainly induced
by the people who remained in the area and that it is not, for example, the quality of
the environment that motivates such moves. Niedomysl and Amcoff (2011) also
found that return migrants are motivated to move to live closer to family and
friends more often than non‐return migrants. Single households also mention
proximity to family and friends more often. It is probably more important for this
group to have their social network nearby. The same is true for the youngest and the
older age groups; people aged 35‐54 mention this motive less often. The youngest
age group can often be in the family‐building phase, during which time being close
to family and friends is likely to be regarded as important (see also Niedomysl and
Amcoff, 2011). From the answers there are some signs that the older age groups
would rather see their children and grandchildren more often, or be able to receive
the support of their children when their age leads to health problems. Finally, the
lowest income group is more likely to be motivated by proximity to family and
friends compared to the highest income group.
In an explorative bivariate analysis for the ‘low house prices’ motivation, we
found the opposite of what would perhaps be expected – that the lowest income
group mentioned this motivation least of all. We also found that highly educated
movers mentioned low house prices more often. However, the logistic regression
analysis proves that income and education are not significant predictors for being
motivated to move to a less‐popular rural area by low house prices. Our dataset
includes owner‐occupiers as well as renters. In the Netherlands the rental prices for
social housing do not vary much between areas. Consequently, renters rarely
mention house prices as a reason for choosing the area. The non‐existent
relationship between income and the low house price motive could be caused by the
fact that the low‐income group is overrepresented in the renting group and therefore
does not mention house prices as a reason for choosing the area. However, when the
77
analysis is repeated with only the owner‐occupiers included, the results are
comparable to those shown in Table 7.
Yet other background characteristics do play a role in being motivated by low
house prices. Single households mention this motivation more often. People moving
from elsewhere in the Netherlands also mention being motivated by low house
prices in the area more often than people moving from surrounding municipalities.
This can probably be explained by the fact that for people moving from further
away, house prices are low in comparison to house prices near their previous place
of residence, while for people moving from nearer by, this difference is less striking.
This also relates to people’s search range. People moving from further away
probably searched in a wide range of rural areas, and in this choice process house
prices apparently played an important role, while people moving from nearby
probably only searched in the surrounding areas, where house prices do not differ
that much between different rural areas. People who had lived in the municipality
previously are less motivated by low house prices, which also applies to people in
the 55‐64 age group compared to the youngest age group.
Finally, work‐related motivations for choosing this specific rural area are
mentioned more often by highly educated movers. This seems surprising at first
sight, because the availability of jobs for which a high level of education is required
does not seem to be a distinctive feature of these less‐popular rural areas. However,
this finding can be explained by the fact that this motivation not only encompasses
living close to work, but also living at an acceptable distance from the job. Previous
research has shown that higher education graduates are more inclined to commute
longer distances (Sandow, 2008). It is probably also the case that, due to the more
limited availability of jobs for people with higher education in or near these areas,
this group only moves once they have already found a job. Movers in the 35‐44 age
group mention work‐related motivations more often than the youngest age group,
while movers over 65 unsurprisingly mention this motivation less often than the
youngest age group. Moreover, people moving from elsewhere in the northern
Netherlands or elsewhere in the Netherlands mention this motivation less often than
people moving from surrounding municipalities. This is contrary to the finding of
78
Halfacree (1994), who found that employment‐related reasons were more important
in long‐distance moves. Apparently, long‐distance movers to these less‐popular
areas are not attracted by the availability of jobs.
3.8 Conclusion and discussion
Our study of migration to less‐popular rural areas in the northern Netherlands
reveals a diverse group of migrants. They are characterised by having a relatively
lower income, a relatively large proportion of people in work, and a large
proportion of highly educated people. They are predominantly young, a substantial
majority are ‘return to the rural’ migrants, and a small proportion moved from an
urban area. The mobility circles vary from local – from surrounding municipalities –
to longer distance – from elsewhere in the Netherlands – but the majority made a
regional move from within the northern Netherlands. Based on migrant
characteristics, we can conclude that our migrants are most similar to the diversity
of migrants to fringe areas in Denmark (Andersen, 2011).
Quality‐of‐life motivations, which form a central part of the traditional
conceptualisation of counterurbanisation, also play a role in decisions to move to
less‐popular rural areas. However, disentangling them into motivations for moving
to a rural area in general and to the specific rural area has proved that these quality‐of‐
life motivations are especially important in the choice for rural living in general. For
moving to the specific rural area, housing characteristics were the main motivation,
directly followed by the physical qualities of the environment, and personal reasons,
such as moving in with a partner and living close to family and friends. Low house
prices in the area only come fifth in the hierarchy of motivations.
Combining the characteristics and motivations of the movers has revealed the
diversity within the movers group. The physical qualities of the environment attract
a group of higher educated movers with high incomes, and middle‐aged people.
Furthermore, our analysis shows a group of movers motivated by living close to
family and friends, consisting of return migrants, singles, the youngest and oldest
age groups and also the lowest income group. Low house prices are an important
79
motivation for young movers, single people and for people moving from elsewhere
in the Netherlands, but not for the low‐income migrants. The housing characteristics
motive refers in most cases to the availability of a specific house and is particularly
mentioned by people moving from an urban area.
As part of the ‘critical re‐appraisal’ of counterurbanisation, it is increasingly
acknowledged that this process is not evenly spread: rural areas differ in their
popularity as places to live (Argent et al., 2007; Bijker and Haartsen, in press;
McGranahan, 2008; Woods, 2011). However, this notion has not yet been sufficiently
translated into research studying the characteristics and motivations of migrants to
rural areas, which has hitherto often focused on attractive, amenity‐rich areas. We
have attempted to fill this gap by studying migration into less‐popular areas. Our
results empirically support the idea that counterurbanisation is indeed a ‘complex
and differentiated phenomenon’ (Woods, 2011: p. 184), also in less‐popular areas. In
particular, our analysis linking motivations to the characteristics of movers has
helped to reveal this diversity. We have seen that less‐popular areas also attract
middle‐class movers motivated by aspects of the rural idyll. However, the results of
our analysis show that less‐popular areas also attract other groups with other
motivations. Our results indicate that in addition to differences in types of rural
areas, the linkages between motivations and the characteristics of migrants are
important to explain migration flows. Therefore, future research in popular rural
areas could also use this method of analysis to further explore the presence of
different groups of migrants with different motivations. Further investigation of the
diversity of migrants to different types of rural areas is required to come to an
empirically underpinned broader conceptualisation of counterurbanisation.
Acknowledgements
The authors would like to thank Hielke Veldkamp, Akje Brouwer, Niels Grootjans
and Daan Vermeer, students of the Faculty of Spatial Sciences of the University of
Groningen, for their contribution to the collection and entry of data. We would also
80
like to thank the municipalities of Ferwerderadiel, De Marne, Menterwolde and
Reiderland for their cooperation.
References
Andersen, H.S., 2011. Explanations for long‐distance counter‐urban migration into
fringe areas in Denmark. Population, Space and Place 17, 627‐641.
Argent, N., Smailes, P., Griffin, T., 2007. The amenity complex: towards a framework
for analysing and predicting the emergence of a multifunctional countryside in
Australia. Geographical Research 45, 217‐232.
Bijker, R.A., Haartsen, T., in press. More than counter‐urbanisation: migration to
popular and less‐popular rural areas in the Netherlands. Population, Space and
Place, doi: 10.1002/psp.687.
Bolton N., Chalkley, B., 1990. The rural population turnaround: a case‐study of
North Devon. Journal of Rural Studies 6, 29‐43.
Deller, S.C., Tsai, T., Marcouiller, D.W., English, D.B.K., 2001. The role of amenities
and quality of life in rural economic growth. American Journal of Agricultural
Economics 83, 352‐365.
Elbersen, B., 2001. Nature on the Doorstep: the Relationship between Protected
Natural Areas and Residential Activity in the European Countryside. Alterra,
Wageningen.
Feijten, P., Hooimeijer, P., Mulder, C.H., 2008. Residential experience and residential
environment choice over the life‐course. Urban Studies 45, 141‐162.
Findlay, A.M., 2005. Editorial: vulnerable spatialities. Population, Space and Place
11, 429‐439.
Findlay, A.M., Short, D., Stockdale, A., 2000. The labour‐market impact of migration
to rural areas. Applied Geography 20, 333‐348.
Fitchen, J.M., 1995. Spatial redistribution of poverty through migration of poor
people to depressed rural communities. Rural Sociology 60, 181‐201.
Foulkes, M., Newbold, K.B., 2008. Poverty catchments: migration, residential
mobility, and population turnover in impoverished rural Illinois communities.
Rural Sociology 73, 440‐462.
81
Gkartzios, M., Scott, M., 2009. Residential mobilities and house building in rural
Ireland: evidence from three case studies. Sociologia Ruralis 50, 64‐84.
Grimsrud, G.M., 2011. How well does the ‘counter‐urbanisation story’ travel to other
countries? The case of Norway. Population, Space and Place 17, 642‐655.
Haartsen, T., Huigen P.P.P., Groote P., 2003. Rural areas in the Netherlands.
Tijdschrift voor Economische en Sociale Geografie 94, 129‐136.
Haartsen, T., Venhorst, V., 2010. Planning for decline: anticipating on population
decline in the Netherlands. Tijdschrift voor Economische en Sociale Geografie
101, 218‐227.
Halfacree, K., 1994. The importance of ‘the rural’ in the constitution of
counterurbanisation: evidence from England in the 1980s. Sociologia Ruralis 34,
164‐189.
Halfacree, K., 2004. A utopian imagination in migration’s terra incognita?
Acknowledging the non‐economic worlds of migration decision‐making.
Population, Space and Place 10, 239‐253.
Halfacree, K., 2008. To revitalise counterurbanisation research? Recognising an
international and fuller picture. Population, Space and Place 14, 479‐495.
Halfacree, K., 2012. Heterolocal identities? Counter‐urbanisation, second homes, and
rural consumption in the era of mobilities. Population, Space and Place 18, 209‐
224.
Halliday, J., Coombes, M., 1995. In search of counterurbanisation: some evidence
from Devon on the relationship between patterns of migration and motivation.
Journal of Rural Studies 11, 433‐446.
Hoggart, K., 2007. The diluted working classes of rural England and Wales. Journal
of Rural Studies 23, 305‐317.
Luttik, J., 2000. The value of trees, water and open space as reflected by house prices
in the Netherlands. Landscape and Urban Planning 48, 161‐167.
Marlet, G., 2009. De Aantrekkelijke Stad. VOC Uitgevers, Nijmegen.
McGranahan, D.A., 2008. Landscape influence on recent rural migration in the U.S.
Landscape and Urban Planning 85, 228‐240.
82
Milbourne, P., 2007. Re‐populating rural studies: migrations, movements and
mobilities. Journal of Rural Studies 23, 381‐386.
Mulder, C.H., 1993. Migration: A Life Course Approach. Thesis, Amsterdam.
Niedomysl, T., Amcoff, J., 2011. Why return migrants return: survey evidence on
motives for internal return migration in Sweden. Population, Space and Place 17,
656‐673.
Niedomysl, T., Malmberg, B., 2009. Do open‐ended survey questions on migration
motives create coder variability problems? Population, Space and Place 15, 79‐87.
Sandow, E., 2008. Commuting behaviour in sparsely populated areas: evidence from
northern Sweden. Journal of Transport Geography 16, 14‐27.
Schudde, B., Gardenier, J.D., Broersma, L., Van Dijk, J., 2010. Noordelijke
Arbeidsmarktverkenning 2010. CAB, Groningen.
Smith, D.P., Phillips, D.A., 2001. Socio‐cultural representations of greentrified
Pennine rurality. Journal of Rural Studies 17, 457‐469.
Smith, D.P., King, R., 2012. Editorial introduction: re‐making migration theory.
Population, Space and Place 18, 127‐133.
Steenbekkers, A., Simon, C., Veldheer, V., 2006. Thuis op het Platteland: De
Leefsituatie van Platteland en Stad Vergeleken. Sociaal en Cultureel Planbureau,
The Hague.
Steenbekkers, A., Simon, C., Vermeij, L., Spreeuwers, W., 2008. Het Platteland van
Alle Nederlanders: Hoe Nederlanders het Platteland Zien en Gebruiken. Sociaal
en Cultureel Planbureau, The Hague.
Stockdale, A., 2006. Migration: Pre‐requisite for rural economic regeneration?
Journal of Rural Studies 22, 354‐366.
Van Dam, F., 2000. Revealed and stated preferences for rural living: evidence from
the Netherlands. In: Haartsen, T., Groote, P., Huigen, P.P.P. (Eds.), Claiming
Rural Identities: Dynamics, Contexts, Policies. Van Gorcum, Assen, pp. 80‐91.
Van Dam, F., Heins, S., Elbersen, B.S., 2002. Lay discourses of the rural and stated
and revealed preferences for rural living: some evidence of the existence of a
rural idyll in the Netherlands. Journal of Rural Studies 18, 461‐476.
83
Visser, P., Van Dam, F., 2006. De Prijs van de Plek: Woonomgeving en Woningprijs.
Nai Uitgevers/Ruimtelijk Planbureau, Rotterdam/The Hague.
Walford, N., 2007. Geographical and geodemographic connections between different
types of small area as the origins and destinations of migrants to Mid‐Wales.
Journal of Rural Studies 23, 318‐331.
Walmsley, D.J., Epps, W.R., Duncan, C.J., 1998. Migration to the New South Wales
North Coast 1986‐1991: lifestyle motivated counterurbanisation. Geoforum 29,
105‐118.
Woods, M., 2011. Rural. Routledge, Oxon/New York.
84
85
4. Different areas, different people? Migration to popular and
less‐popular rural areas in the Netherlands4
Abstract
Counterurbanisation is often conceptualised as urban, middle‐class movers attracted by an
idyllic rural setting. However, knowing that rural areas differ in their popularity for living.
we argue that more attention is required to the diversity of rural in‐migration within
countries. We do so by comparing the characteristics, motivations and values of movers to
popular and less‐popular areas in the northern Netherlands using multivariate analysis on
survey data (N = 1717). In contrast to earlier studies we focus on the motivations for
choosing the destination area instead of mixing those with motives for leaving. We also
included residential history and values in our analysis. The differences between movers to
the two types of areas appear to be less distinct than indicated by previous studies.
Although popular areas more often attract middle‐class movers, both areas attract urban
movers. Movers to less‐popular areas are more often motivated by low house prices and
moving in with a partner, but movers to popular areas also mention instrumental
considerations related to work and location. While the physical aspects of the environment
are important to moving to popular areas, social aspects are more important to moving to
less‐popular areas. We find that values add to our understanding of counterurbanisation. It
is not possible to relate two distinct groups of movers directly to different types of rural
areas within countries. Future research into rural migration should be careful not to use too
simplified understandings of counterurbanisation, both in general and with regard to
different types of rural areas.
4.1 Introduction
In recent years the dominant conceptualisation of counterurbanisation as the
movement of middle‐class groups from the city in search of a new life in an idyllic
4 This chapter is reprinted with some minor adjustments from: Bijker RA, Haartsen T, Strijker D.
Different areas, different people? Migration to popular and less‐popular rural areas in the
Netherlands. Population, Space and Place. DOI: 10.1002/psp.1741. (available online in ‘early view’)
Copyright © 2012 John Wiley & Sons, Ltd.
86
rural setting (Halfacree, 2008), has been reconsidered critically in several papers,
including some contributions in this journal (e.g. Bijker and Haartsen, 2012;
Grimsrud, 2011; Halfacree, 2001, 2008, 2012; Milbourne, 2007). One aspect of this
reconsideration is the exploration of the international dimension of the
counterurbanisation concept. Grimsrud (2011) shows that caution is required when
applying this dominant conceptualisation originating in the core regions of Europe
and America to remote rural areas. Her study in Norway shows movers to rural
areas as motivated by family relations and economic concerns, instead of a desire for
a rural lifestyle. Nonetheless, Bijker and Haartsen (2012) argue that in addition to
these differences between countries, attention is also required to the diversity of
rural in‐migration within countries, even in densely populated and urbanised
countries like the Netherlands. This follows from what Woods (2005) refers to as the
‘regionally uneven nature of counterurbanization’, the finding that rural areas differ
in their popularity for rural living. The popularity of rural areas for migration
purposes is found to be mainly influenced by their accessibility and the presence of
amenities (Argent et al., 2007; Bollman and Briggs, 1992; Johnson and Beale, 1994;
McGranahan, 2008). The results of Bijker and Haartsen (2012), based on migration
statistics and secondary data, indicated that also within the core regions in Europe
the classical counterurbanisation model does not explain all rural population
change. While the characteristics of the movers to popular rural areas in the
northern Netherlands fitted the counterurbanisation story, less‐popular rural areas
in the same region shared personal reasons as an important motive for in‐migration
with more remote rural areas in Europe (see for example Grimsrud, 2011).
Bijker and Haartsen (2012) conclude by stating that a more refined
understanding of the phenomenon of counterurbanisation can be achieved by
further considering the question of whether different types of rural areas attract
different groups of migrants. They recommend separating the motivations for
location choice from the motives for leaving the previous residence, instead of the
analysis they performed, in which these two types of motivations were mixed. They
also felt it important to pay more attention to the residential history of movers.
Collecting our own survey data in seven municipalities in the northern Netherlands
87
(N = 1717) allowed us to incorporate these recommendations in this study.
Therefore, in our current analysis we focus specifically on the motivations for
choosing the destination area. We also include return migration and incorporate a
more detailed description of the previous residence of the movers. In addition,
besides characterising the movers using ‘traditional’ socio‐demographic
characteristics, we also use people’s values. These underlying motivational factors
have been used in housing research to improve the understanding of housing
preferences (e.g. Coolen and Hoekstra, 2001; Coolen et al., 2002). In migration
research values have received little attention so far – we want to explore whether
they can also aid in distinguishing movers to different types of rural areas.
We include the values along with background characteristics and motivations
in a multivariate analysis, which allows us to investigate which of these variables
predict a move to a specific type of rural area, while controlling for the effect of
other factors. Following Bijker and Haartsen (2012), we differentiate rural areas
based on differences in popularity for living. We believe that such a direct
comparison of migration flows to popular and less‐popular areas can offer better
insight into the diversity of rural population change within countries, compared to
studies focusing on one type of rural area, which are the majority within research
into rural migration. Only a few studies directly compare migration flows to
different types of rural areas (Gkartzios and Scott, 2009; Grimsrud, 2011; Hjort and
Malmberg, 2006), but these studies most often use univariate analyses. The results of
the multivariate analysis are discussed after a short overview of the literature on
rural in‐migration and the use of values in housing and migration research.
4.2 Migration to rural areas
As stated earlier, rural in‐migration is often defined as counterurbanisation
(Champion, 1989; Boyle and Halfacree, 1998), mostly referring to the movement of
middle‐class families from cities attracted by the quality of the residential
environment (Halfacree, 2008). Following Halfacree (1994), physical and social
features of the residential environment can be identified. Physical features are for
88
example space, fresh air and an attractive landscape, while social features include
friendly people, peaceful living and less crime (Gkartzios and Scott, 2009; Halfacree,
1994). The quality of the residential environment as a motive for moving to the rural
is often connected to the existence of the rural idyll: the pull of the countryside as a
way of life (Benson and O’Reilly, 2009). Elements of this positive representation of
the countryside are a less hurried lifestyle, peace and quiet, space and greenness:
this idyllic countryside also offers an escape from hectic urban life (Cloke, 2003;
Short, 2006; Van Dam et al., 2002).
This view of counterurbanisation is generally based on research into rural
areas that are popular to live in. Studying migration to less‐popular areas has helped
bring some nuance to this classical image. However, the number of studies
investigating migration to rural areas that can be regarded as less popular to live in
is small. With regard to the background characteristics of movers, the limited
evidence so far from different national contexts reveals the movers to be different
from the middle‐class movers in the counterurbanisation model. In a study of
depopulating, peripheral rural areas in Scotland, Stockdale showed few in‐migrants
to be employed in professional or managerial sectors and only half of them to
possess tertiary or vocational qualifications. Impoverished rural areas in the US with
a wide availability of cheap rental housing attract large groups of low‐income
movers (Fitchen, 1995; Foulkes and Newbold, 2008). However, similar to in‐
migrants to Danish fringe areas experiencing population decline and a low level of
economic activity (Andersen, 2011), a more diverse group of movers was found in a
study focusing on less‐popular areas in the northern Netherlands with respect to
income and education (Bijker et al., in press). Nonetheless, compared to movers to
rural areas in general in the Netherlands, their level of income appeared to be
relatively low. Furthermore, using migration statistics and limited secondary data,
Bijker and Haartsen (2012) found in the same region that compared to popular areas,
less‐popular areas attracted more migrants without higher education. With respect
to the origin and age of the movers, it appears that less‐popular areas attract young
movers who have often made a local or regional move, coming from cities as well as
rural areas (Bijker and Haartsen, 2012; Bijker et al., in press; Stockdale, 2006).
89
Looking at the motivations of movers to less‐popular areas, it appears that
instead of the quality of the residential environment, other motivations for choosing
the destination area are more central in the decision process. Studies have shown the
importance of more personal motivations, such as marriage and the proximity of
family and friends, employment considerations and housing reasons, in some cases
in particular related to the availability of affordable housing (Fitchen, 1995; Foulkes
and Newbold, 2008; Grimsrud, 2011; Stockdale, 2006). Also in the previously
mentioned study in the northern Netherlands, it appeared that the motive of living
close to family and friends was more important for movers to less‐popular areas
(Bijker and Haartsen, 2012).
However, other studies show that the rural character of the destination is not
totally insignificant to movers to less‐popular areas. Foulkes and Newbold (2008)
found that the low‐income movers to depressed rural communities in the US also
saw their move as an opportunity to satisfy the desire to live in a rural setting. In
less‐popular rural areas in the northern Netherlands quality‐of‐life motivations were
particularly important in choosing rural living in general. However, when choosing
a particular rural area to live in, the physical qualities of the environment were also
an important motivation, in addition to housing characteristics, more personal
considerations and the low house prices in the area (Bijker et al., in press).
The previous overview already shows that counterurbanisation is a more
diverse phenomenon than the more narrow conceptualisation often used in
research. Halfacree (2008, 2012) has attempted to develop a more general and
flexible model, which includes a broader range of people and experiences than is
typically the case. The model focuses on the motivations of the rural in‐migrants,
regardless of the origin of their move, be it an urban or rural area. Three groups of
migrants are distinguished based on the extent to which ‘the pull of rurality’
motivated the migration. For the first group, labeled ‘mainstream
counterurbanisation’, the rural environment is important, balanced by more
practical considerations, such as proximity to employment or services. These are the
people often represented in the ‘classical’ counterurbanisation concept. For the
90
second group, ‘back‐to‐the‐land counterurbanisation’, the pull of rurality is
absolutely central. This group has a separate position in the model because of their
post‐migration lifestyle, which is radically different from the mainstream group (see
for example Halfacree, 2001). The third group is labeled ‘default
counterurbanisation’: a group for which the rural character of the place is almost
totally incidental and for which instrumental considerations – for example relating
to employment or family support – are more important (Halfacree, 2008).
Halfacree (2008) suggests using this flexible model to explore ‘the ways in
which counterurbanisation is constructed and deployed in a variety of contexts’ (p.
492). We want to investigate to what extent the different groups in the model are
connected to different rural contexts within countries. Based on the findings from
studies focusing on one type of rural area described above, it can be concluded that
migration to popular rural areas is often defined as ‘mainstream
counterurbanisation’, while migration to less‐popular areas is often seen as ‘default
counterurbanisation’. In this study we want to find out whether this distinction
remains when migration to these different types of rural areas is directly compared.
In his discussion of the model, Halfacree (2008) also states that ‘the underlying
motivations’ of counterurbanisation, which may at least be partly unconscious for
the person involved, seem until now to be only very partially addressed in research.
The inclusion of the values people have in our analysis may be an attempt to gain
more insight in these underlying motivations.
4.3 Adding values to the analysis
In recent years in housing research the idea has developed that traditional
background characteristics (i.e. income, education or age) may no longer be
sufficient to explain residential preferences due to a trend of greater differentiation
in housing behaviour (see for example Heijs et al., 2009, 2011; Jansen, 2011). It is
argued that this differentiation is generated by several demographic, socioeconomic
and sociocultural changes that have taken place in Western countries, such as
smaller households and a larger variation in household types, a greater variety of
91
specific lifestyle‐based subcultures and the increased share of affluent households
(Kersloot and Kauko, 2004). This has resulted in approaches which explore the
motives underlying consumers’ preferences, in order to improve the understanding
of housing preferences and preferences for residential environments. In these
approaches consumers are seen as acting in a goal‐oriented manner, choosing a
particular dwelling to satisfy values and goals that are important to them (Bettman,
1979; Jansen, 2011; Kersloot and Kauko, 2004; Rokeach, 1973). The concepts ‘values’
and ‘goals’ are closely related, as Schwartz (1994) defines values as ‘desirable
transsituational goals, varying in importance, that serve as guiding principles in the
life of a person or other social entity’ (p. 21). Values are thus seen as objectives
which, consciously or unconsciously, act as criteria in all our actions (Rokeach,
1973).
An example of research into the relationship between values and housing
preferences is the small‐scale study of Coolen and Hoekstra (2001), in which they
use semi‐structured interviews to investigate the values underlying preferred
housing attributes. It appears that, for example, the preference for a garden can be
based on various values, such as freedom, unity with nature, enjoying life and
creativity. Using a quantitative approach, Coolen et al. (2002) study values as
determinants of intended tenure choice. They use regression analysis to investigate
the role that values play in the choice between renting and owning, when other
important characteristics such as income and age are controlled for. They find the
value orientation ‘power and achievement’ to be related to ownership, while ‘family
values’ are related to renting. Coolen et al. conclude that the value orientation
‘power and achievement’ is related to viewing a house as a capital good, while the
‘family values orientation’ is more related to a house being seen as a consumer good.
In our study we explore whether including value orientations in our analysis
helps characterise movers to different types of rural areas. In migration research,
underlying motivational factors such as values have been paid little attention. An
exception is Christenson’s study (1979), which investigated the values of
metropolitan and nonmetropolitan stayers and potential migrants in North Carolina.
It appeared that potential migrants from nonmetropolitan areas attached more
92
importance to the values of personal freedom, individualism and racial and sexual
equality than did nonmetropolitan stayers. He found no comparable differences
between metropolitan stayers and potential migrants.
This limited attention in recent research to the values underlying migration
decisions is rather strange, given that they seem to fit very well with the increasing
attention being paid to the cultural dimension of migration in migration research
(Halfacree, 2004). As Fielding stated: ‘Migration tends to expose one’s personality, it
expresses one’s loyalties and reveals one’s values and attachments (often previously
hidden). It is a statement of an individual’s world‐view, and is, therefore, an
extremely cultural event’ (1992, p. 201). This statement expresses that migration
decision‐making is more than a question of satisfying housing and employment
requirements (Lewis, 1998). Individuals identify with locations, including their
residential location, and give them social meaning. Choosing a certain kind of
dwelling in a certain kind of place is an expression of a person’s lifestyle, taste and
identity (De Wijs‐Mulkens, 1999; Van der Horst et al., 2002).
4.4 Methodology
The data we present in this paper were collected in popular and less‐popular rural
areas in the northern Netherlands. In comparison with other European countries, the
Netherlands is densely populated and urbanised, though a considerable part of the
country is nonetheless defined as rural by the Dutch population (Haartsen et al.,
2003). In this study we focus on the northern Netherlands, being the most rural part
of the country, both based on address density and the perception of the Dutch
people (Haartsen et al., 2003). Following our previous paper (Bijker and Haartsen,
2012) we operationalise the popularity for living of the rural areas in our study using
the average house prices per municipality. House prices reflect the value buyers
attach to houses and their surroundings (Luttik, 2000; Visser and Van Dam, 2006).
This average house price incorporates the residential environment as well as
housing characteristics such as size and quality. The use of house prices instead of
in‐migration numbers, which are often used in studies of rural areas’ popularity
93
(Argent et al., 2007; McGranahan, 2008), is more appropriate to the Dutch context. In
the Netherlands the development of new housing is strictly regulated by the
government and building is only allowed in a limited number of areas. These
restrictions lead to price increases rather than increased in‐migration when there is
increasing demand (Marlet, 2009). Rental prices differ little between regions in the
Netherlands, which is why these are not considered in defining the popularity of
rural areas for this study.
Based on house prices, we distinguish three types of rural areas in the northern
Netherlands: less‐popular rural areas, average rural areas and popular rural areas
(Bijker and Haartsen, 2012). These areas are considered rural based on Statistics
Netherlands’s national standard for urbanity, the so‐called address density,
consisting of municipalities with on average less than 1,000 addresses per square
kilometre. In this study we focus on the less‐popular and popular rural areas.
Figure 1 shows that the areas with the lowest house prices, the less‐popular
areas, are located along the northern and eastern borders of the northern
Netherlands. They can be characterised as having an open, marine clay landscape
with a large share of large‐scale arable farmland. Compared to the popular areas,
they have relatively fewer natural and recreational areas and they have a relatively
low share of hotel and catering industry employment. They are relatively distant
from the centre of the Netherlands and the people that inhabit these areas have a
lower average household income than those in the average and popular rural areas.
In these less‐popular areas, population decline and a decline in the number of
households are both expected (Bijker and Haartsen, 2012), and in some
municipalities population decline is already taking place (Haartsen and Venhorst,
2010). Popular rural areas, on the other hand, can be characterised as having a more
semi‐open landscape with predominantly grazing livestock farming, a landscape
type that is generally perceived as attractive (McGranahan, 2008; Ulrich, 1986). They
also have a relatively large share of natural areas. Popular rural areas have an
expected household growth and are the least distant from the centre of the
Netherlands (Bijker and Haartsen, 2012).
94
Figure 1. Three types of rural areas in the northern Netherlands
In the less‐popular rural areas, we selected four municipalities as study areas:
De Marne, Menterwolde and Reiderland in the province of Groningen, and
Ferwerderadiel in the province of Friesland. In the popular areas we selected three
municipalities: Aa en Hunze and Westerveld in the province of Drenthe, and
Wymbritseradiel in the province of Friesland. (see Figure 1). None of these
municipalities have a large town within their boundaries. Therefore, in the Dutch
context all migrants to these municipalities can be considered to be rural migrants,
rather than movers to a larger settlement.
In cooperation with the municipalities, we selected all the households who had
moved into the municipality from another municipality in the period 2005–2009
using the population register. Owner‐occupiers as well as renters were included in
this selection. The relatively short period of 5 years was chosen to reduce the risk of
memory‐recall problems of the respondents and in particular post hoc rationalisation
(see e.g. Walmsley et al., 1998). In autumn 2009 a postal questionnaire was sent to all
95
the households that met the selection criteria in the four less‐popular municipalities.
In an accompanying letter we asked whether a household member aged over 18
could complete the questionnaire. In the summer of 2010 this procedure was
repeated in the three popular municipalities. A total of 1717 (664 for the less‐popular
areas, 1053 for the popular areas) questionnaires were returned, representing a
response rate of 24%. In the Netherlands this is considered to be a good response
rate for a postal survey. Nevertheless, if the non‐response is selective, it affects the
representativeness of the outcomes. However, this is not so much a problem when
investigating relationships between variables like we do in this paper. There is no
reason to assume that these relationships differ for people inside and outside the
sample. Furthermore, there are no indications that the pattern of non‐response
differs between the different types of rural areas.
The questionnaire covered a variety of topics: residential history, demographic
and socioeconomic characteristics, housing characteristics, previous experience with
the area and the motivations that had been important in the migration decision.
Following Van Dam (2000) and Bolton and Chalkley (1990), we separated the
motives for leaving the previous residence from the motives for choosing the
destination area. Regarding the motives for choosing the destination area, we
distinguished between choosing to live in a rural area in general and choosing to
live in the specific less‐popular or popular rural area.
We chose to use open‐ended questions to investigate the motivations for
moving to a rural area in general and for choosing the specific rural area under
study, to take into account the multidimensionality of the motives for moving.
Open‐ended questions enable a more open‐minded and less biased approach to
studying migration motives (Halfacree, 2004; Niedomysl and Malmberg, 2009). The
respondents could offer more than one motivation and afterwards every answer was
categorised. The categories were based on previous studies of rural migration (e.g.
Bolton and Chalkley, 1990; Gkartzios and Scott, 2009; Halfacree, 1994; Halliday and
Coombes, 1995; Walmsley et al., 1998) and also on what emerged from the survey
96
results. Niedomysl and Malmberg (2009) have shown that coder variability when
coding open‐ended questions on migration motives is relatively low.
The questionnaire also included 8 items measuring value orientations. These
items stem from a questionnaire used previously in a study in the Netherlands (see
Coolen et al., 2002). This original questionnaire was based on the operationalisation
of ten value domains by Schwartz (1992). Nevertheless, in the Coolen et al. Dutch
study only eight value domains could be retrieved from the data: Basic values,
Hedonism, Family values, Structure and order, Power and achievement, Self‐esteem,
Esteem from others and Self‐actualization. In the original questionnaire four items
were included for every value domain (Coolen et al., 2002). However, because
adding 32 items to our questionnaire would have made this too long, we chose to
use one item for every value domain. We tried to select the item that covered the
value domain best in the context of our study. While we are aware that using a
single item results in a less reliable measurement compared to a scale consisting of
several items measuring a construct, we believe that using one item per value can
help us explore the additional value of these underlying motivational factors for
distinguishing movers to different types of rural areas. The respondents were asked
to rate the importance to their lives of each value on a five‐point scale (with the
categories: very unimportant, unimportant, averagely important, important, very
important).
4.5 Results
We used logistic regression analysis to compare the movers to popular and less‐
popular rural areas in the northern Netherlands. The regression model estimates
how various background characteristics, the ten most mentioned motivations for
choosing to live in the area in question and eight personal values influence the type
of rural area people moved to. To determine the relevance of the significant effects,
the distribution of these independent variables is shown in Table 1. The logistic
regression model omits all cases with missing values for any of the independent
variables, which resulted in a total of 1345 respondents included in the model (844
97
movers to popular areas, 501 movers to less‐popular areas). The dependent variable
in the model consists of the categories ‘having moved to a popular rural area’ (value
0) and ‘having moved to a less‐popular area’ (value 1). The three groups of
independent variables were added to the model in three steps to see their added
explanational value to the model and to analyse how they influence the relationship
between the independent variables already included in the model and the
dependent variable. The results of the analysis are presented in Table 2. A positive B
coefficient means that increasing the value of the independent variable increases the
probability of having moved to a less‐popular rural area. If the sign of the B
coefficient is negative, an increased value on the independent variable increases the
probability of having moved to a popular rural area. We checked for
multicollinearity in all three models, there appeared to be no problem.
Table 1. Descriptive statistics of the independent variables for the respondents included in
the logistic regression model for popular rural areas (PA) and less‐popular rural areas (LPA)
in the northern Netherlands (N = 1345).
Background characteristics (%) PA LPA Values (1‐5 scale, average) PA LPA
Moved from urban area 48 44 Freedom 4.53 4.56
Lived in municipality before 14 13 Pleasure in life 4.54 4.57
Household income (net per month)*** A harmonious family life 4.37 4.40
< EUR 1250 euros 7 11 Self‐discipline 3.72 3.76
EUR 1250‐2500 34 44 Wealth 2.94 2.89
EUR 2500‐4000 38 35 Being rational 3.91 3.96
> EUR 4000 21 10 Preserving public image* 2.43 2.53
Household composition Varied life*** 3.43 3.58
Couple without children 52 49
Couple/single parent with children 33 35 Motivation for choosing this area (%) PA LPA
Single household 15 16 Physical qualities*** 35 24
Previous residence*** Housing characteristics 23 26
Surrounding municipalities 24 26 Work‐related*** 18 12
Elsewhere in the northern Netherlands 28 36 Living close to family and friends 15 12
Elsewhere in the Netherlands 45 33 Location** 12 8
Abroad 3 5 Quietness 11 9
Higher education *** 57 46 Familiarity with the area** 11 8
Age*** Low house price*** 5 14
20‐34 20 32 Social qualities 8 9
35‐44 23 21 Moving in with partner*** 5 13
45‐54 17 18
55‐64 23 20
> 65 18 9
* p<0.10; ** p<0.05, *** p<0.01 (Chi2‐test, T‐test for values)
98
Model 1, only including the background variables, shows that households
earning more than EUR 2500 net per month move to popular rural areas more often
than the lowest income group (< EUR 1250 per month). This is an effect that would
be expected, considering that the two types of areas are distinguished based on
house price. However, when controlled for income, there are other background
characteristics which have a significant effect on the type of area people move to.
Highly educated movers more often move to popular rural areas. Combined with
the effect of income it appears that the popular areas attract a group of movers that
fit the ‘classical’ middle‐class image of counterurbanisation. This corresponds with
the findings of our earlier study based on secondary data (Bijker and Haartsen,
2012). Nevertheless, that study also found that people moving from urban areas
more often moved to popular rural areas, another aspect of the classical view on
counterurbanisation. However, in our current analysis people moving from urban
areas showed the same probability of moving to popular or to less‐popular areas.
Apparently, including more detailed information on the residential history of the
respondents, namely a variable measuring return migration (‘lived in the
municipality before’) and a more detailed variable on the previous residence,
removes the effect of having moved from an urban area. Indeed, also in our current
research it appeared that when return migration was excluded from the analysis,
‘having moved from an urban area’ again becomes significant.
Regarding this residential history, it appears that people who have lived in the
same municipality before have a greater probability to have moved to a popular
area, meaning that the popular rural areas attract a greater share of return migrants.
We expected familiarity with the area to be more important for less‐popular areas,
based on the assumption that the less‐popular areas have a less positive
representation due to their more peripheral position in the Netherlands and a less
attractive landscape (see also Bijker and Haartsen, 2012) and the finding in earlier
research that familiarity with an area creates more positive representations
(Haartsen, 2002). Nevertheless, this positive effect of familiarity with an area
naturally also applies to popular areas and apparently has an even stronger effect
for these areas.
99
Table 2. Logistic regression analysis of moving to less‐popular vs. popular rural areas
Model 1 Model 2 Model 3
Background characteristics B B B
Moved from urban area (urban = 1, rural
= 0)
‐0.194 ‐0.192 ‐0.190
Lived in municipality before (yes = 1, no
= 0)
‐0.336* ‐0.232 ‐0.236
Income (ref. < EUR 1250)
EUR 1250‐2500 ‐0.257 ‐0.238 ‐0.193
EUR 2500‐4000 ‐0.625*** ‐0.538** ‐0.499**
> 4000 euros ‐1.275*** ‐1.187*** ‐1.197***
Household composition
(ref. couple without children)
Couple/single parent with children ‐0.021 0.020 0.034
Single household ‐0.112 ‐0.021 ‐0.021
Previous residence (ref. surrounding
municipalities)
Elsewhere in the northern Netherlands 0.434*** 0.415** 0.393**
Elsewhere in the Netherlands ‐0.031 ‐0.111 ‐0.155
Abroad 0.785** 0.708** 0.559
Higher education (yes = 1, no = 0) ‐0.280** ‐0.186 ‐0.220
Age (ref. > 65)
20‐34 1.193*** 0.994*** 0.998***
35‐44 0.747*** 0.728*** 0.740***
45‐54 0.728*** 0.728*** 0.733***
55‐64 0.579*** 0.624*** 0.637***
Motivation for choosing the area
Physical qualities ‐ ‐0.445*** ‐0.494***
Housing characteristics ‐ 0.079 0.087
Living close to family and friends ‐ ‐0.205 ‐0.226
Work‐related ‐ ‐0.540*** ‐0.539***
Quietness ‐ ‐0.082 ‐0.063
Familiarity of the area ‐ ‐0.303 ‐0.309
Location ‐ ‐0.397* ‐0.404*
Moving in with partner ‐ 0.774*** 0.798***
Low house price ‐ 0.962*** 0.990***
Social qualities ‐ 0.395* 0.428*
Values
Freedom ‐ ‐ 0.120
Pleasure in life ‐ ‐ ‐0.214*
A harmonious family life ‐ ‐ 0.102
Self‐discipline ‐ ‐ ‐0.045
Wealth ‐ ‐ ‐0.130
Being rational ‐ ‐ 0.190**
Preserving public image ‐ ‐ 0.065
Varied life ‐ ‐ 0.156**
Constant 0.577 0.473 ‐0.436
N 1345 1345 1345
Nagelkerke R2 0.103 0.165 0.180
2 105.421*** 172.835*** 190.396***
p < 0.10; ** p < 0.05, *** p < 0.01
100
The more recent residential history is included in the model by means of the
previous residence of the respondents. It appears that people who moved from
elsewhere in the northern Netherlands compared with people who moved from
surrounding municipalities more often have moved to less‐popular areas. People
who moved from abroad – consisting of Dutch people returning to the Netherlands
and foreign in‐migrants – were also more likely to have moved to less‐popular areas.
However, this is only a very small group within the total group of movers (see Table
1). Other studies (Stockdale, 2006) found less‐popular rural areas to mainly attract
local movers, but our multivariate analysis showed popular and less‐popular areas
attract movers from elsewhere in the Netherlands to the same extent.
Finally, we find the movers to less‐popular areas to be younger than those who
moved to popular areas. All the age groups have a higher probability to move to
less‐popular areas compared to the oldest age group (> 65), with the youngest age
group (aged 20‐34) having the strongest effect. This corresponds with earlier
findings in the northern Netherlands (Bijker and Haartsen, 2012; Bijker et al., in
press) and in less‐popular areas in other countries (Stockdale, 2006).
Subsequently, the ten most mentioned motivations for choosing the specific
rural area were added to the background characteristics in the model (model 2).
After this was done, the Nagelkerke R2 increased substantially, indicating that the
motivations add a lot of explanatory value to the model. Physical qualities refer to
answers such as nature, space, the presence of water and beautiful surroundings.
People who are motivated by these physical qualities of the environment are more
likely to have moved to popular rural areas. This fits with the image of the
counterurbanite attracted by ‘an idyllic rural setting’ (Halfacree, 2008). A possible
explanation for the importance of this motive for moving to popular areas can be
found in the attractiveness of the semi‐open landscape and the presence of nature in
these areas.
People who mention work‐related motivations (i.e. ‘close to work’, ‘acceptable
commuting distance’ and ‘started my own business’) for choosing the area are also
more likely to have moved to popular areas. This could partly be explained by their
location closer to the central part of the Netherlands. Indeed, people who mention
101
the location of the area (i.e. ‘central location’, ‘not too far from the Randstad area’
and ‘close to the city of Groningen’) as an important motivation are also more likely
to have moved to popular rural areas.
On the other hand, people who mention that they moved to the area to live
with their partner are more likely to have moved to less‐popular areas. People who
mention low house prices as a motivation for choosing the area are also more likely
to have moved to less‐popular areas. The importance of these two motivations
corresponds with the findings of studies focused exclusively on less‐popular areas
(Fitchen, 1995; Foulkes and Newbold, 2008; Stockdale, 2006).
A perhaps more surprising finding is that the social qualities of the
environment – such as freedom, friendliness, nice atmosphere, lots of things going
on in the village and the mentality of the people – are a more important motivation
for people who have moved to less‐popular areas compared to movers to popular
areas. This is in contrast to the popular areas, which have the physical qualities of
the environment as an important pull factor. It shows that the rural character of an
area does matter for movers to less‐popular areas. Apparently it is useful to
distinguish between the physical and social aspects of the rural to explain moving to
different types of rural areas. It also raises the question of whether – similarly to
how the attractiveness of the landscape in popular areas may be linked to the
importance of the motivation of the physical qualities of the environment – less‐
popular areas may have more to offer with regard to the social qualities of an area.
Unfortunately, it is not possible to answer this question within the scope of this
study. It nonetheless indicates that more attention is required to ‘social amenities’, in
addition to the physical characteristics most often used to capture the amenity of
rural areas (Argent et al., 2007; McGranahan, 2008).
The other motivations included in the model do not have a significant effect on
moving to a popular or a less‐popular area. Based on the findings of Bijker and
Haartsen (2012), which corresponded to the outcomes in other studies (Fitchen,
1995; Foulkes and Newbold, 2008; Grimsrud, 2011), we expected that living close to
family and friends would be a more important motivation for movers to less‐
popular areas. However, when including only the motivations for choosing the
102
destination area in the analysis instead of mixing the motivations for leaving the
previous residence and choosing the destination, it appears that the importance of
this motivation does not differ for movers to popular and less‐popular areas.
Most background characteristics already included in the model remain
significant after adding the motivations. Nevertheless, for ‘lived in the municipality
before’ and ‘higher education’ the significant effect disappears. Because of this
finding we explored possible interaction effects between these background
characteristics and the motivations. An interaction effect occurs when the effect of
one independent variable on the dependent variable depends on the level of a
second independent variable. However, also in the additional logistic regression
analysis (results not shown in the paper) including these interaction effects, ‘lived in
the municipality before’ and ‘higher education’ remained not significant. This
finding indicates that the results are robust for the inclusion of interaction effects.
In the final, third step, eight personal values were added to the model (model
3). The Nagelkerke R2 increased again, meaning that the values do indeed add
explanatory value to the model. Even though it is a small addition, the contribution
of the value orientations to the model comes on top of the characteristics and
motivations of movers often used when studying migration into rural areas. The
finding that values add less to the explanatory power of the model than adding
motivations was confirmed when both were added to the model in reverse order
(results are not shown here).
Three value orientations have a significant effect. The more important people
think it is to be rational in their lives (in the questionnaire with the addition in
brackets ‘intelligent, logical, thoughtful’), the higher the probability that they moved
to less‐popular rural areas. These areas probably offer a rational choice to these
movers, for example they enable buying a larger house for less money. The more
importance people attach to variation in their life, the higher the probability they
moved to a less‐popular area. This value refers to ‘a life full of challenge, novelty
and change’. Apparently the less‐popular areas offer a kind of adventurous choice
for these movers. On the other hand, the more importance people attach to pleasure
103
in life (with the addition ‘enjoyment, satisfaction’), the greater the probability that
they moved to a popular rural area. It seems that for these movers, popular areas
represent places to enjoy life.
After adding the value orientations to the model, all the background
characteristics and motivations that had a significant influence on the type of rural
area remained significant, except for people who had previously lived abroad. This
shows that the value orientations have an independent effect on the type of area
people moved to, without interacting with the background characteristics and
motivations.
4.6 Conclusion
In this paper we compared the characteristics, motivations and values of movers to
popular and less‐popular rural areas in the northern Netherlands using multivariate
analysis. This study follows a previous study in the same region using migration
statistics and secondary data, which results indicated that attention is required to the
diversity of rural in‐migration within countries. While the characteristics of the
movers to popular rural areas in that study fitted the counterurbanisation story very
well, less‐popular rural areas shared personal reasons as an important motive for in‐
migration with more remote rural areas in Europe (Bijker and Haartsen, 2012). We
further investigated the question of whether different types of rural areas within
countries attract different migrants with different motivations, using survey data
which allows for a more comprehensive analysis. Whereas in the previous study
motivations for location choice were mixed with motives for moving, in our current
analysis we were able to focus on motivations for choosing the destination area. We
were also able to include return migration in our analysis and a more detailed
measurement of the previous residence of the movers. In addition to characterising
the movers using ‘traditional’ socio‐demographic characteristics, we also used
people’s values to explore whether these underlying motivational factors have an
added value for distinguishing movers to different types of rural areas.
104
When directly comparing the movers to popular and less‐popular areas it
appears that movers to popular areas more often belong to higher income groups
and are more often higher educated. They also more often have lived in the
municipality before and more often belong to the oldest age cohort. The physical
qualities of the environment, work‐related considerations and the location of the
area are more often mentioned by them as motivations to move to an area.
Compared to movers to less‐popular areas, they attach more importance to the value
of ‘pleasure in life’. The movers to less‐popular areas are more likely to have made a
move from elsewhere in the northern Netherlands and they are more likely to
belong to the younger age groups. Compared to the movers to popular areas they
more often mentioned moving in with their partner and the low house prices in an
area as motivations to move there. Movers to less‐popular areas less often mention
the physical qualities of the environment, but in contrast they more often mention
social qualities. Compared to movers to popular areas they attach more importance
to the values of ‘being rational’ and ‘varied life’.
These results show that when using this more comprehensive multivariate
analysis to compare movers to popular and less‐popular areas, the differences
between the two groups are less distinct then previously argued by Bijker and
Haartsen (2012) and as indicated by studies focusing on one of the types of rural
area. Although our current analysis does show that popular rural areas attract
middle‐class movers more often, it appears that both areas attract movers from
urban areas. When considering motivations, movers to less‐popular areas are more
motivated by low house prices and the personal motivation of moving in with a
partner, which suggests that moving to less‐popular areas can indeed be defined as
‘default counterurbanisation’ (see Halfacree, 2008). However, the finding that
movers to popular areas are more often motivated by instrumental considerations
such as work‐related reasons and the location of the area than movers to less‐
popular areas and that the motivation to live close to family and friends is of the
same importance to movers to either type of area, suggests that popular areas also
attract a group of ‘default counterurbanisers’. Moreover, when identifying the
105
physical and social aspects of the rural character of the environment, which is the
main motivation for the category of ‘mainstream counterurbanisation’ (see
Halfacree, 2008), it appears that while the physical aspects are important for moving
to popular areas, social aspects are more important for moving to less‐popular areas.
Apparently, the rural character of the destination also matters for those moving to
less‐popular areas.
Based on these outcomes we can conclude that studying in‐migration to
different types of rural areas within a country shows that counterurbanisation does
indeed entail more than what the ‘classical’ conceptualisation would suggest. Our
results thus empirically support the broader conceptualisation of
counterurbanisation in the model developed by Halfacree (2008). However, our
results also show that the two main groups distinguished in the model, mainstream
and default counterurbanisation, are not directly related to different types of rural
areas within countries. Future research into rural migration should be careful not to
use too simplified understandings of counterurbanisation, both in general and with
regard to different types of rural areas within countries.
Finally, it is interesting to see that values can add to the understanding of
migration processes. Even though their addition is small, they contribute further to
the characteristics and motivations of migrants when comparing movers to different
types of rural areas. Moreover, they appear to have a direct effect on the type of area
people move to, instead of interacting with or having an effect through the
motivations for choosing an area, which could also be expected. This probably
shows that these values are more related to more abstract ideas or representations
that people have of the areas, which are not captured in the more concrete
motivations. Apparently, the popular areas are regarded as places to enjoy life,
while the less‐popular areas represent a rational choice or even a kind of
adventurous choice. In addition to this, the importance of the ‘varied life’ value for
movers to less‐popular areas could indicate the existence of some kind of group of
creative people moving to these – in the Dutch context – peripheral areas (e.g. Bell
and Jayne, 2010; Gibson, 2010). In this study we were only able to measure the
values with a single item, resulting in a less reliable measurement. Based on our
106
results it would appear worthwhile to further explore the additional value of the use
of values in migration research, using scales including more items to measure value
orientations.
Acknowledgements
The authors would like to thank Henny Coolen for providing the value
questionnaire and Hielke Veldkamp, Akje Brouwer, Niels Grootjans and Daan
Vermeer for their role in collecting the data.
References
Andersen HS. 2011. Explanations for long‐distance counter‐urban migration into
fringe areas in Denmark. Population, Space and Place 17: 627‐641.
Argent N, Smailes P, Griffin T. 2007. The amenity complex: towards a framework for
analysing and predicting the emergence of a multifunctional countryside in
Australia. Geographical Research 45: 217‐232.
Bell D, Jayne, M. 2010. The creative countryside: policy and practice in the UK rural
cultural economy. Journal of Rural Studies 26: 209‐218.
Benson M, O’Reilly K. 2009. Migration and the search for a better way of life: a
critical exploration of lifestyle migration. The Sociological Review 57: 608‐625.
Bettman JR. 1979. An Information Processing Theory of Consumer Choice. Addison‐
Wesley: Reading.
Bijker, RA, Haartsen T. 2012. More than counter‐urbanisation: migration to popular
and less‐popular rural areas in the Netherlands. Population, Space and Place 18:
643‐657.
Bijker RA, Haartsen T, Strijker, D. in press. Migration to less‐popular rural areas in
the Netherlands: exploring the motivations. Journal of Rural Studies. DOI:
10.1016/j.jrurstud.2012.07.003.
Bollman RD, Briggs B. 1992. Rural and small town Canada: an overview. In Rural
and Small Town Canada, Bollman RD (ed.). Thompson Educational Publishing:
Toronto.
107
Bolton N, Chalkley B. 1990. The rural population turnround: a case‐study of North
Devon. Journal of Rural Studies 6: 29‐43.
Boyle P, Halfacree K. 1998. Migration into Rural Areas: Theories and Issues. John Wiley
& Sons: Chichester.
Champion AG (ed.). 1989. Counterurbanization: the Changing Pace and Nature of
Population Deconcentration. Edward Arnold: London.
Christenson J. 1979. Value orientations of potential migrants and nonmigrants. Rural
Sociology 44: 331‐344.
Cloke PJ. 2003. Country Visions. Pearson Education: Harlow.
Coolen H, Hoekstra J. 2001. Values as determinants of preferences for housing
attributes. Journal of Housing and the Built Environment 16: 285‐306.
Coolen H, Boelhouwer P, Van Driel K. 2002. Values and goals as determinants of
intended tenure choice. Journal of Housing and the Built Environment 17: 215‐236.
De Wijs‐Mulkens E. 1999. Wonen op Stand: Lifestyles en Landschappen van de Culturele
en Economische Elite. Het Spinhuis: Amsterdam.
Fielding T. 1992. Migration and culture. In Migration Processes and Patterns (volume 1):
Research Progress and Prospects, Champion T, Fielding T (eds.). Belhaven Press:
London/New York; 201‐212.
Fitchen JM. 1995. Spatial redistribution of poverty through migration of poor people
to depressed rural communities. Rural Sociology 60: 181‐201.
Foulkes M, Newbold KB. 2008. Poverty catchments: migration, residential mobility,
and population turnover in impoverished rural Illinois communities. Rural
Sociology 73: 440‐462.
Gibson C. 2010. Guest editorial – Creative geographies: tales from the ‘margins’.
Australian Geographer 41: 1‐10.
Gkartzios M, Scott M. 2009. Residential mobilities and house building in rural
Ireland: evidence from three case studies. Sociologia Ruralis 50: 64‐84.
Grimsrud GM. 2011. How well does the ‘counter‐urbanisation story’ travel to other
countries? The case of Norway. Population, Space and Place 17: 642‐655
Haartsen T. 2002. Platteland: Boerenland, Natuurterrein of Beleidsveld? Een Onderzoek
naar Veranderingen in Functies, Eigendom en Representaties van het Nederlandse
108
platteland. Koninklijk Nederlands Aardrijkskundig Genootschap/Rijksuniversiteit
Groningen: Utrecht/Groningen.
Haartsen T, Huigen PPP, Groote P. 2003. Rural areas in the Netherlands. Tijdschrift
voor Economische en Sociale Geografie 94: 129‐136.
Haartsen T, Venhorst V. 2010. Planning for decline: anticipating on population
decline in the Netherlands. Tijdschrift voor Economische en Sociale Geografie 101:
218‐227.
Halfacree K. 1994. The importance of ‘the rural’ in the constitution of
counterurbanization: evidence from England in the 1980s. Sociologia Ruralis 34:
164‐189.
Halfacree K. 2001. Constructing the object: taxonomic practices,
‘counterurbanisation’ and positioning marginal rural settlement. International
Journal of Population Geography 7: 395‐411.
Halfacree K. 2004. An utopian imagination in migration’s terra incognita?
Acknowledging the non‐economic worlds of migration decision‐making.
Population, Space and Place 10: 239‐253.
Halfacree K. 2008. To revitalise counterurbanization research? Recognising an
international and fuller picture. Population, Space and Place 14: 479‐495.
Halfacree K. 2012. Heterolocal identities? Counter‐urbanisation, second homes, and
rural consumption in the era of mobilities. Population, Space and Place 18: 209‐224.
Halliday J, Coombes M. 1995. In search of counterurbanisation: some evidence from
Devon on the relationship between patterns of migration and motivation. Journal
of Rural Studies 11: 433‐446.
Heijs W, Carton M, Smeets J, Van Gemert A. 2009. The labyrinth of life‐styles. Journal
of Housing and the Built Environment 24: 347‐356.
Heijs W, Van Deursen A, Leussink M, Smeets J. 2011. Re‐searching the labyrinth of
life‐styles. Journal of Housing and the Built Environment 26: 411‐425.
Hjort S, Malmberg G. 2006. The attraction of the rural: characteristics of rural
migrants in Sweden. Scottish Geographical Journal 122: 55‐75.
109
Jansen SJT. 2011. Lifestyle method. In The Measurement and Analysis of Housing
Preference and Choice, Jansen SJT, Coolen HCCH, Goetgeluk RW (eds.). Springer:
Dordrecht; 177‐202.
Johnson KM, Beale CL. 1994. The recent revival of widespread population growth in
nonmetropolitan areas of the United States. Rural Sociology 59: 655‐667.
Kersloot J, Kauko T. 2004. Measurement of housing preferences: a comparison of
research activity in the Netherlands and Finland. Nordic Journal of Surveying and
Real Estate Research 1: 144‐163.
Lewis GJ. 1998. Rural migration and demographic change. In The Geography of Rural
Change, Ilbery B (ed.). Pearson Education Limited: Harlow; 131‐160.
Luttik J. 2000. The value of trees, water and open space as reflected by house prices
in the Netherlands. Landscape and Urban Planning 48: 161‐167.
Marlet G. 2009. De Aantrekkelijke Stad. VOC Uitgevers: Nijmegen.
McGranahan DA. 2008. Landscape influence on recent rural migration in the U.S.
Landscape and Urban Planning 85: 228‐240.
Milbourne P. 2007. Re‐populating rural studies: migrations, movements and
mobilities. Journal of Rural Studies 23: 381‐386.
Niedomysl T, Malmberg B. 2009. Do open‐ended survey questions on migration
motives create coder variability problems? Population, Space and Place 15: 79‐87.
Rokeach MJ. 1973. The Nature of Human Values. Free Press: New York.
Schwartz SH. 1992. Universals in the content and structure of values: theoretical
advances and empirical tests in 20 countries. In Advances in Experimental Social
Psychology, Vol. 25, Zanna MP (ed.). Academic Press: San Diego; 1‐65.
Schwartz SH. 1994. Are there universal aspects in the structure and contents of
human values? Journal of Social Issues 50: 19‐45.
Short B. 2006. Idyllic ruralities. In Handbook of Rural Studies, Cloke PJ, Marsden T
(eds.). Sage: London; 133‐148.
Stockdale A. 2006. Migration: Pre‐requisite for rural economic regeneration? Journal
of Rural Studies 22: 354‐366.
Ulrich RS. 1986. Human responses to vegetations and landscapes. Landscape and
Urban Planning 13: 29‐44.
110
Van Dam F. 2000. Revealed and stated preferences for rural living: evidence from
the Netherlands. In Claiming Rural Identities: Dynamics, Contexts, Policies,
Haartsen T, Groote P, Huigen PPP (eds.). Van Gorcum: Assen; 80‐91.
Van Dam F, Heins S, Elbersen BS. 2002. Lay discourses of the rural and stated and
revealed preferences for rural living: some evidence of the existence of a rural
idyll in the Netherlands. Journal of Rural Studies 18: 461‐476.
Van der Horst H, Kullberg J, Deben L. 2002. Wat Wijken Maakt: De Wording van
Functionele, Sociale en Expressieve Kwaliteiten van Vreewijk, Zuidwijk en Ommoord.
DGVH/Nethur: Utrecht.
Visser P, Van Dam F. 2006. De Prijs van de Plek: Woonomgeving en Woningprijs. Nai
Uitgevers/Ruimtelijk Planbureau: Rotterdam/Den Haag.
Walmsley DJ, Epps WR, Duncan CJ. 1998. Migration to the New South Wales North
Coast 1986‐1991: Lifestyle motivated counterurbanisation. Geoforum 29: 105‐118.
Woods M. 2005. Rural Geography. Sage Publications: London.
111
5. Using a diary approach to explore the residential search
process of movers to rural areas5
Abstract
This paper investigates how people find new places to live in rural areas in the northern
Netherlands. We use a diary approach, a new method for studying the residential search
process. Our findings show that the search areas of local, regional and distant searchers
differ in scale and in the extent to which they change during the search process.
Explanations for these different search patterns include the local ties of local searchers
and the social and work‐related ties that restrict the search range of distant searchers.
Contact with an area, resulting in positive perceptions, appears to be essential for
including an area in the search space. All searchers also mentioned areas they avoided
based on perceived characteristics, most often social characteristics. Using a diary
approach provides more insight into the non‐linear character of the search process and
the less tangible aspects that play a role.
5.1 Introduction
In the context of the transition of the rural from a production to a consumption
space, in‐migration is often seen as an important opportunity for the development of
rural areas (e.g. Andersen, 2011; Stockdale, 2006). Many studies have investigated
the characteristics and motivations of potential and actual migrants into rural areas,
often focusing either on people considering a move to the rural (e.g. Van Dam et al.,
2002) or on people who have already made a move into the rural (e.g. Bijker and
Haartsen, 2012; Bijker et al., in press; Gkartzios and Scott, 2009; Grimsrud, 2011).
Little attention has been paid to the process between stated intentions to move and
actual moving behaviour in a rural context.
5 This chapter is reprinted from: Bijker, RA, Haartsen, T, Strijker D. Using a diary approach to explore
the residential search process of movers to rural areas and has been submitted to an international
journal.
112
The first aim of this paper is to gain more insight into the residential search
process in rural areas. We distinguish local and non‐local searchers to determine to
what extent their residential search process differs. This could help clarify why non‐
local movers are less likely to realise their rural location preferences than local
movers (De Groot et al., 2012). This is a relevant question in the context of the
processes of population decline that have started in many rural areas in Europe
(Haartsen and Venhorst, 2010). Furthermore, our own work on migration into rural
areas has yielded questions related to the residential search process. For instance, to
what extent can the finding that more distant movers to rural areas in the northern
Netherlands are more often motivated by low house prices be explained by the
different search ranges of distant and local movers? (Bijker et al., in press). Does
familiarity with an area create more positive representations and increase the
chances of moving or at least searching there? (see also Bijker and Haartsen, 2012).
The second aim of the paper is to explore the usefulness of a new method for
studying the search process: a diary approach with both qualitative and quantitative
elements. Until now, the search process has been mainly studied using retrospective
survey methods, computer experiments, retrospective interviews and simulation
models (e.g. Donaldson, 1973; Hooijmeijer and Oskamp, 1996; Phipps, 1983;
McPeake, 1998; Harper, 1991). There are few examples of longitudinal research into
the search process (Huff, 1986; Cronin, 1982; Goetgeluk, 1997). A diary approach in
which people are repeatedly approached during their search has the potential to
offer insight into what happens between the longer intervals of a longitudinal
approach. Furthermore, it prevents the risk of post hoc rationalisation that is a
feature of using retrospective methods (see e.g. Walmsley et al., 1998). Finally, by
including qualitative elements in the approach, it offers more insight into the often
overlooked, more emotional, non‐economic, less tangible aspects of human decision‐
making (see e.g. Levy et al., 2008). Our respondents were all looking for houses in
the northern Netherlands. We recruited them through real estate agents in the area.
We chose to locate our study in the northern Netherlands as the most rural part of
the country, based both on address density and the perceptions of the Dutch people
(Haartsen et al., 2003). Furthermore, the North contains a large variety of rural areas
113
in terms of characteristics such as house prices, landscape, accessibility, building
style and history (see also Bijker and Haartsen, 2012), which makes the region very
suitable for studying search behaviour.
We start our paper by discussing the literature on residential search in general
and in rural areas more specifically. We pay some specific attention to the concept of
representations. The diary approach is explained in the methodology section,
followed by the results and conclusion.
5.2 Theory and previous research
5.2.1 The residential search process
The individual decision‐making process with respect to residential mobility includes
the formation of a positive attitude towards moving, the search for and evaluation of
housing alternatives and ultimately the decision to move or to stay (De Groot et al.,
2012). If intending movers are unable to realize their preferences, they may choose to
stay in their current homes (Brown and Moore, 1970) or resort to substitution: the
acceptance of a new home that may satisfy some but not all of their initial
preferences (Goetgeluk, 1997). The realization of intentions to move depends on the
interaction between the triggers or motives for moving, housing preferences,
individual resources and restrictions on the one hand, and the opportunities and
constraints in the housing market on the other (De Groot et al., 2011a; Hooimeijer
and Oskamp, 1996). Often, the motive for moving develops from one of the careers
that together form the life course: the residential, the household, education or
employment career (Mulder, 1993). Anticipated changes in these careers can also
affect the probability that the intention to move is realized because some triggers for
moving are more urgent than others. While intended moves triggered by changes in
the household, employment or educational careers are usually associated with a
high degree of necessity, moves triggered by residential motives are often less
urgent (Goetgeluk, 1997). On the other hand, unanticipated household and
employment changes can lead to the postponement or cancellation of the intended
114
move or result in an unexpected move within a short period (De Groot et al., 2011b;
Kan, 1999; Speare, 1974).
The residential choice process occurs within the context of the housing market.
By searching, the potential migrant interacts with the market. The concept of search
includes several interrelated characteristics: it is a goal‐directed activity, it involves a
complex process of information gathering; a point is reached where search ends and
a choice is made; and it happens in a context of uncertainty and within a set of
constraints (Clark and Flowerdew, 1982). The process of search may be
characterized by its duration, the type of information sources used, the number of
houses examined, and the radius of the area searched (Huff, 1982; Walmsley and
Lewis, 1993).
Brown and Moore (1970) developed a model for residential search behaviour in
which they include the influence of information on the spatial aspects of search. In
this model, the potential migrant possesses a spatially selective image of the area
based on information acquired from previous experiences prior to starting the
search. The term ‘awareness space’ is used to refer to those locations within the
overall space about which the intended migrant household has knowledge before
the search starts, based on regular, direct contact; or based on indirect contact, such
as through acquaintances’ experiences and the mass media (Brown and Moore, 1970;
Walmsley and Lewis, 1993). Based on the preferences of the household within this
awareness space a ‘search space’ is defined. A preference for any particular place
would not depend on the objective measures of that place, but on the filtered
information which is the basis of the individual’s cognitive environment (Voets,
1994). For example, newspapers might present houses for sale or to let from an
entire area, but many such adverts will not be considered, either because they are
not located in neighbourhoods known by the intended migrant or because they are
located in areas considered unlikely to contain acceptable properties (Brown and
Moore, 1970). From the search space, the decision‐maker constructs a choice set: a set
of alternatives to be considered more carefully (Voets, 1994). In another approach to
choice, the choice assumed to take place is between accepting or not accepting
particular housing opportunities in sequence (Mulder, 1996).
115
In addition to the household’s existing knowledge of opportunities or
locations, potential migrants use different information sources during the search
process, including newspapers, walking or driving around, friends and relatives,
and real estate agents (e.g. Barrett, 1976; Rossi, 1955; Walmsley and Lewis, 1993).
More recently, the internet has appeared on this list. The limited evidence so far
shows that newcomers to an area are more likely to use the internet, and searchers
using the internet visited a larger number of houses personally, contrary to what
was expected (Palm and Danis, 2002). The internet enables rural migrants to search
across the whole country for houses based on a set of search criteria, rather than
searching in particular regions (Niedomysl, 2010). However, it appears that the
internet has had little impact on search patterns (Niedomysl, 2010; Palm and Danis,
2002).
Information plays an important role in the search process. After an inspection
of a property, the household should have a sufficient impression of what Koopman
(2012) calls observable attributes, such as the dwelling itself, accessibility, physical
features and even the area’s atmosphere. Nonetheless, what is difficult to determine
is the social quality of an area, the ‘“soft” socioeconomic and demographic attributes
that shape the intangible, dynamic and largely unobservable social quality of the
area’ (Koopman, 2012, p. 35), which also includes the frequency and nature of social
interactions among residents. This is something that should be experienced rather
than observed, which is why residents have an information advantage in this
respect. House‐seekers can adopt several strategies to avoid making the wrong
choice. Using spatial search strategies (see Huff, 1986) can help reduce the amount of
information that needs to be collected on distant neighbourhoods. Another strategy
is relying on the neighbourhood’s reputation (Koopman, 2012), a concept which is
discussed in the next section.
The search process is not static or linear, it changes over time and can be seen
as a learning process. As the migrant gains experience of the area during his search,
some areas may come to be included in the search space, while other areas may be
eliminated (Brown and Moore, 1970; Walmsley and Lewis, 1993). According to
Goetgeluk (1997), during their search process, people will also learn about the
116
opportunities and constraints to achieving their housing preferences. During the
search process, their preferences will therefore change and become more realistic in
relation to the housing market.
5.2.2 The role of representations in the search process
As mentioned in the previous section, the delimitation of the search space within the
awareness space and the choice set within the search space is based on ‘the filtered
information which is the basis of the individual’s cognitive environment’ (Voets,
1994, p. 12). It is assumed that people base their behaviour on their image,
interpretation or representation of geographical space (e.g. Bunce, 1994; Halfacree,
1994; Short, 1991; Walmsley et al., 1998). For rural areas in general the existence of
the ‘rural idyll’ is often seen as an explanation for the decision to move to the rural
(Boyle and Halfacree, 1998; Halfacree, 1994; Van Dam et al., 2002). Moscovici (1982)
defines representations as ‘cognitive systems of preconceptions, images and values
which have their own cultural meaning’ (p. 12). Social representations exist in and
are produced by communication and interaction (e.g. Holloway and Hubbard, 2001).
Two forms of communication are important in this: mass communication and
interpersonal communication (Haartsen, 2002). Direct experience with the object of
representation is also important. People who are familiar with the countryside
because they have lived there or visit it frequently have other and more positive
representations of the rural (Haartsen, 2002; Van Dam et al., 2002). Our assumption
is that also representations of specific rural places influence the residential search
and decision process and vice versa.
A concept related to representations used in housing research in an urban
context is that of neighbourhood reputation. Neighbourhood reputation can be
defined as ‘the meaning and assessment assigned by residents and outsiders to the
neighbourhood.’ (Hortulanus, 1995, p. 42). This definition implies that a reputation
is an image shared by a significant number of individuals; nevertheless, the same
neighbourhood can have different reputations for different groups of people.
Reputations differ for residents and non‐residents, the internal reputation with
residents is often more positive and more detailed. It appears that the rating of a
117
neighbourhood is most strongly influenced by the socioeconomic and ethnic
composition of the neighbourhood, physical and functional characteristics appear to
be less important (Permentier, 2009). As discussed in the previous section, relying on
a neighbourhood’s reputation during the search process can be a risk‐minimising
strategy when information about the social quality of an area is lacking (Koopman,
2012).
5.2.3 The search process in rural areas
Few studies have investigated the residential search process in a rural context. Lewis
and Sherwood (1994, as cited in Lewis, 1998) found in England that households with
locational ties to a district had usually made their choice of district well before the
decision to migrate was activated. Those households without a connection to their
ultimate chosen district tended to consider around eight possibilities, using a variety
of information sources, though the final choice was mostly based on personal
experience. For both groups, however, the choice of village or small town involved a
tendency to focus the search on part of a district, and the availability of a particular
type of house was the most significant criterion. This seems to be contrary to
Harper’s finding (1991) that settlement selection is increasingly replacing property‐
based selection. She signals that instead of being based on personal and prior
association, the decisions of the movers were often based on abstract preconceptions
of the area (cf. Walmsley et al., 1998). Niedomysl (2010) also found that in Sweden
only few migrants considered more than two regions to move to, but most
considered a number of alternative locations within an area.
5.3 Methodology
We used a diary approach to study the residential search process in rural areas. This
is a new method to study the residential search process, but has been used
previously in other scientific fields and topics (e.g. Richardson, 1994; Laurenceau
and Bolger, 2005; Larson, 1989; Sudman and Ferber, 1971; Rieman, 1993). Richardson
(1994) identifies the ability to obtain data about processes as an advantage of the
118
diary approach, capturing the time sequence of events and perceptions as they
unfold, rather than relying on recall of past events. Traditional longitudinal designs
can also address these questions, but because they typically involve only a small
number of repeated measurements taken at long intervals, they cannot capture
changes in the same way (Bolger et al., 2003). The advantages of a diary approach fit
very well with the idea that the search for a new house is a learning process, in
which the search space, representations of areas and preferences are likely to change
over time.
5.3.1 Diary approach
Our respondents were looking for houses in the northern Netherlands. We started
the diary approach with a semi‐structured in‐depth interview with each respondent.
This was done because respondents were recruited at different stages of their search
process. The interview was used to obtain information about the search process to
that point. However, the interview also offered an opportunity for meeting the
respondents in person. Personal interactions are regarded as important for the
successful application of a diary approach, because they can help motivate the
respondents (Richardson, 1994; Rieman, 1993). In addition to the search process, the
questions considered some background characteristics, the residential history of the
household and the extent of contact with rural areas in the North prior to starting
the search. Furthermore, respondents were asked about their motives for moving
and their residential preferences. We used a map with regions in the northern
Netherlands to further discuss the search area and the associations that came up
when thinking about the search area and the surrounding areas. Some of the
respondents shared their search process with the interviewer by providing some
examples of their favourite houses. One of the respondents produced a roadmap of
the Netherlands on which he had marked his search areas. Most of the interviews
were conducted at the respondents’ homes, but for practical reasons, in three cases
the respondent’s workplace was used for the interview. Some of interviews were
done with both of the partners, some with only one. It was left to the respondents to
119
decide which of the two options they preferred. The interviews were recorded and
later they were transcribed and analysed.
Whereas a time‐based design requires participants to report on their
experiences at regular intervals or in response to a signal given by a signalling
device, event‐contingent studies require participants to provide a self‐report each
time the event in question occurs. When using an event‐based design, it is important
that events are easy to identify for respondents (Bolger et al., 2003). The residential
search process consists of easy and less easy to define events. Moreover, it requires a
lot of effort on the part of respondents for them to remember to complete the
questionnaire after a relevant event has occurred over a longer period. Therefore, we
chose a time‐based design to study the residential search process.
Consequently, every two weeks after the interview until they bought a house
or quit searching, the respondents were invited by e‐mail to complete an electronic
questionnaire by clicking on a link. When using a fixed time schedule for a diary
approach, the length of intervals is an important consideration (Bolger et al., 2003). It
seems that the search process can change pace and intensity. Therefore, the interval
should not be too long: in an intensive period things could be missed or forgotten.
To avoid a research design that is too demanding for the respondents, the interval
should not be too short either. Inviting the respondents by e‐mail every two weeks
reminds them to complete the questionnaire, just like a signalling device (Bolger et
al., 2003). A diary which is short and easy to complete helps keep people motivated
to participate (Richardson, 1994; Stone et al., 1991). The electronic questionnaire was
easy to complete and only showed questions relevant to the respondent, based on
their answers to previous questions. The questionnaire consisted of both closed and
open questions. The questionnaire focused on tracking change in the search process.
When search activities or changes in the process occurred, additional questions
seeking more detail about this event became visible. This design led to a short
questionnaire when little had happened in the preceding two weeks, while more
questions appeared when changes had taken place. Respondents were asked
whether they had bought a house or had quit searching each time. Furthermore,
questions were included about housing preferences and preferences about the
120
residential environment. Respondents could see what they had filled in the last time
and could make changes if their preferences changed. If they did, they were asked
why. The same applied to the search area. Furthermore, they were asked which
search activities they had undertaken, with additional information being asked for
some activities.
The risk of using a diary approach is that completing the diary could influence
the process studied, for example by generating new thoughts about the search
process. People could also feel ‘lazy’ if the diary arrived in their mailbox and they
had not developed any search activities, causing them to feel obliged to try some
new search activity for the sake of it. Therefore, during the explanation of the
method it was emphasised that ‘nothing happened’ in the search process over two
weeks was an acceptable response. So far, there is little evidence of behaviour
change as a result of participating in a diary approach (Bolger et al., 2003; Litt et al.,
1998). ‘Habituation’ could lessen the risk of influencing the process: people get used
to the rhythm of receiving and completing the questionnaire, thus attending to it less
self‐consciously (Bolger et al., 2003). On the other hand, habituation, and more
specifically the development of a habitual response style when completing the diary,
could have negative effects, for instance by causing the development of a tendency
to skim over questions (Bolger et al., 2003).
5.3.2 Recruitment strategy
We recruited the respondents through real estate agents in the northern
Netherlands. The main eligibility criterion was that they were looking for a house in
rural areas in the northern Netherlands. Rural was defined very broadly, including
both villages and surrounding areas. It was quite difficult to find respondents
because of the stagnation in the housing market at the time. We followed several
strategies to approach respondents through the participating real estate agents. We
asked real estate agents to include a message about the study to prospective house
buyers to whom information about new houses on the market was regularly sent, to
which some respondents reacted. If an agent had no such mailing list, we asked
them to include a letter about the research in the information pack that was given to
121
people visiting a house with them. In some cases we received contact details from
people who had visited a house with the real estate agent, which permitted us to
contact them directly to ask whether they wanted to participate. Because we
expected the difference between local and more distant movers to be important to
the search process, we tried to assemble a group of searchers over varying distances.
5.3.3 Study population
The research material provided by the ten respondents offers in‐depth, detailed
information about the residential search process in rural areas of different types of
searchers. Four of the respondents were living outside the northern Netherlands at
the time of recruitment: this group will be termed distant searchers in the remainder
of this paper. Three respondents were already living in the northern Netherlands
and were initially searching within a wider area in the northern Netherlands
(regional searchers). Three of the respondents were also already living in the North
and were initially searching very locally, in the village they were living in or in a
neighbouring village (local searchers). Table 1 presents the characteristics of the
three groups of respondents.
Table 1. Characteristics of the respondents
Local searchers Regional searchers Distant searchers
Age (min.‐max.) 25‐30 27‐55 51‐63
Household composition Couples (one with
children living at
home)
Couples (one with
children living at
home)
Couples (without
children living at
home)
Level of education (min‐
max.)
Secondary‐Higher Secondary–Higher Secondary‐Higher
Max. house price (min‐max.,
EUR)
130,000–280,000 155,000–450,000 200,000‐350,000
122
5.4 Results
5.4.1 Overview of search process
The results section starts with an overview of the search processes of the
respondents, based on the information gathered in the interviews and from the diary
data. After that we focus on some specific aspects of the search process.
Table 2 shows that the search duration varied greatly, from 8 months to 8 years
(while still searching). However, there are no clear differences in this respect
between the three groups of searchers. The reason for moving for all local searchers
relates purely to housing characteristics. For the regional and distant searchers this
is more varied: housing‐related reasons, wanting to leave the current residential
environment, the desire to make a new start, marriage, and the wish to live closer to
a specific school type. Most of the moves were not of a very urgent nature. For some
of the respondents, the urgency was diminished by the fact that they first wanted to
sell their existing houses. Two of the distant searchers clearly indicated that because
their move was not a necessity, the new house should offer a real improvement: one
of them even spoke of a ‘dream house’.
Half the movers changed their house or residential environment preferences
during their searches, in accordance with the idea that searching can be regarded as
a learning process about the possibilities in the housing market (Goetgeluk, 1997).
Nevertheless, Table 2 shows that it is not just about learning about possibilities and
adapting personal preferences to them, sometimes it is also about learning what one
really finds important. For instance, D10 increasingly realized how much she valued
the proximity of family and friends, which led her to search closer to her place of
residence. Most changes in preferences were recorded in the diary questionnaire,
indicating that most changes were made later in the search process. Originally, we
assumed that in order to be able to stay in their preferred place of residence, local
searchers would alter their preferences more, while distant movers would more
readily switch to another search area to fulfil their wishes. However, this
assumption is not confirmed by our data. Both local and regional and distant
movers sometimes adapted their preferences. In addition to changing preferences,
123
Table 2 shows that changes in the household situation can also affect the search
process. L3 decided to quit searching due to her pregnancy and unemployment of
her partner, while financial problems caused R5 initially to switch from a preference
for buying to renting. With respect to the search areas and their development, a clear
difference can be seen between local searchers on the one hand and regional and
distant searchers on the other (see Table 2). Therefore, the next section covers
changes in the search area. Figure 1 shows the places and areas mentioned in
relation to the search areas in Table 2 and the main text.
124
Figure 1. Areas and places mentioned in the description of the search areas of the respondents
Table 2. Overview of search process of respondents (L = local searcher, R = regional searcher and D = distant searcher)
Resp.
nr.
Search
duration
(months)
Reason for
moving
Urgency of the
move
Initial search area
Development of search area
End of search
Changes in
preferences
L1
24Move from
owning to
renting
Desire to m
ove
in one year (at
time of
interview, after
one year
searching)
Their current
village (H
allum)
and neighbouring
villages
No chan
ges
Bought a house
in H
allum
Size of garden could be
less w
ith a sem
i‐
detached house;
building period
importan
t (after 1990);
preference for new
ly
built, but municipality
has no plans
L2
8
Desire for larger
house and a
quiet street
No tim
e limit
for the search to
end
The larger
neighbouring
village (Stiens)
No chan
ges
Bought a house
in Stiens
No
L3
24Move from
renting to
owning and a
larger space
outside
No tim
e limit
Their current
village (H
allum)
No chan
ges
Quit searching
due to
pregnan
cy and
unem
ployment
of partner
No
R4
36 (still
searching)
House had
become too
large, to m
ake a
new start
No tim
e limit,
importan
t to sell
the current
house for a good
price
Northern
Drenthe,
Westerkwartier
Added northwestern
Groningen (around the
current place of residence)
Developed a preference
for three villages:
Zuidlaren, R
oden and
Leek
Still searching,
not expecting to
move in short
term
; not sold
their own house
No
R5
48
Marriage
They w
ant to
live together;
urgency is quite
high
The southeast of
Drenthe, 15 km
around
Hoogeveen,
Oldam
bt, Zeeland
Oldam
bt an
d Zeeland
disap
peared
Focus on area around
Hoogeveen
Added H
oogeveen itself
Bought a house
in H
oogeveen
Switched from
preference for buying
to renting to buying
again; p
reference for a
rural setting, sw
itched
to urban area
R6
15
To live closer to
a specific school
type
Waiting for
more clarity
about the
schools and for
own house to be
sold
Assen and
surroundings
Added Emmen and
surroundings an
d Delfzijl
and surroundings
Focus on ‘10 km around
Delfzijl’
Quit searching
becau
se they
have not sold
their own house
No
D7
14
Job chan
ge,
desire to m
ake a
new start
Desire to m
ove
within a year (at
time of the
interview, after
eight months of
searching), new
house should be
improvem
ent
Veenkoloniën and
Westerw
olde
For a while also around
Emmen, becau
se of the
possibility of a job there
Bought a house
in N
ieuwe
Pekela
Building period of
house less im
portan
t,
preference for older
house rem
ains;
availability of public
tran
sport less
importan
t
D8
29
Desire for larger
house and a
garden, d
esire to
leave the current
place of
residence in the
Ran
dstad area
(lack of open
space/asocial
behaviour)
No tim
e limit,
searching for a
‘dream house’
The North and
Southwest of
Drenthe, also a
small area in the
‘Green H
eart’ in
the Ran
dstad area
Added northern and
northeastern Groningen
Added northern Friesland
and an area in
southeastern Friesland
Focus on northern
Friesland and
northwestern Groningen
Added another area in
southeastern Friesland,
removed it again
Focus on three villages in
northwestern Groningen
and northern Friesland
Focus on northwestern
Groningen and two
villages in particular
Bought a house
in Pieterburen
Preference for a woody
landscap
e, due to high
house prices there they
mad
e a chan
ge to
search areas w
ith an
open lan
dscap
e type;
design of the house and
building period m
ore
importan
t
(‘characteristic’ house);
lower m
axim
um price;
proxim
ity of larger
place and presence of
facilities in the village
less importan
t; size of
the village matters
more
D9
96 (still
searching)
Increasing
inconvenience
caused by
Every year they
say to each
other: ‘this year
In Zeeland (near
their current place
of residence) on
Westerw
olde
Added northeast
Groningen
Still searching,
planning to
move this year
No
people living in
the
neighbourhood
(noise, rubbish)
it is really going
to hap
pen’
the island of
Walcheren
Added northern
Groningen
Focus on the municipality
‘De Marne’
Search area is w
idened
again to Groningen and
Friesland
D10
18
Having to leave
their official
residence due to
retirement of
partner
Move within 18
months at tim
e
of interview (6
months after
beginning the
search)
Veluwe, their
current village in
the Ran
dstad area,
Alblasserwaard,
Betuwe,
Achterhoek,
Amerongen,
northern
Netherlands with
a focus on
northern
Groningen and
southeastern
Friesland
Focus on the northern
provinces
Groningen disap
peared
Betuwe ad
ded again
Widened to the
Netherlands (excep
t
Groningen)
Focus on southern
Friesland, D
renthe an
d
Overijssel
Focus on the
Alblasserwaard
Steenwijk and
surroundings
Alblasserwaard and
Veluwe
Veluwe, Betuwe an
d
around Steenwijk
Alblasserwaard
Overijssel, G
elderland,
Alblasserwaard , close to
current house,
Veenwouden
South of Zwolle, an area
in N
orth‐Braban
t an
d
around N
unspeet
(Veluwe)
Bought a house
in N
unspeet
(Veluwe)
Preference for an older
house, n
ewly‐built now
also an option;
maxim
um price
increased; proximity of
a suitable parish m
ore
importan
t than initially
thought; proximity of
job opportunities more
importan
t; proxim
ity of
family and friends
more importan
t;
importan
t to feel it
‘click’ with the house,
yet the house bought
did not provide that
feeling (from the ’70s,
not an initial
preference)
128
5.4.2 Search area
The search areas of local, regional and distant searchers differed in their scale and
the extent to which they changed during the search process. While local movers
spoke of the villages they were searching in, regional and distant movers initially
spoke about areas, mostly consisting of several municipalities. Furthermore, the
local movers in our study started searching in their own or a neighbouring
village and continued to do that throughout the search process. For the regional
and distant searchers, the process was much less linear. However, the degree of
change varied. In some cases the search area moved to another area, in some
cases the search area was expanded or areas disappeared from the search range.
Nonetheless, some of the regional and distant searchers developed a preference
for a municipality or even specific villages during their search process. In two
cases the ultimate house was also found in these preferred villages. Respondents
mentioned these kinds of changes during their initial interviews, but the diary
questionnaire helped track the changes later in the search process.
It appears that the importance of local ties for local searchers is an
important reason for this distinct difference in search patterns between local and
regional and distant searchers. In the interviews the local searchers mentioned
several types of local ties. Firstly, these ties were related to work or the proximity
of family and friends, and more in general the depth of their roots in a social
network. L2 described the effects of moving to another area as follows:
Then youʹll miss your connection with everything youʹve got here,
then youʹll have to make a completely new start.
Another aspect was the continuation of daily activities, for example not wanting
to change children’s primary schools. Finally, some degree of emotional
attachment to a place played a role. Again, as L2 noted:
Of course there are very many beautiful spots, only when you come
back here you realize this place also has its beauty, we never really felt
like moving far away, I guess.
129
This also refers to a kind of natural attachment to the place that emerged from the
local searchers’ answers. The regional searchers also mentioned local ties that
bound them in their search. In contrast to the local searchers, these were not
restricted to a specific village, as for them it was sufficient to live at ‘a reasonable
distance’.
Whereas these local ties are by definition very area‐specific, most of the
regional and distant searchers were motivated by preferences not confined to one
specific rural municipality. Some of the distant and regional searchers were
attracted by a specific housing type in a rural setting, others also mentioned the
general benefits of rural living and the desire for an attractive landscape in the
vicinity. Their less strong preference for a specific rural location made it easier for
them to change search area. However, as remarked above, the specificity of the
search area also increased during the search process of most regional and distant
movers. It appears that more area‐specific preferences are developed during the
search process.
5.4.3 The role of representations
As discussed in Section 5.2, searchers start with an awareness space based on
direct and indirect contact, and from within this awareness space a search area is
selected (Brown and Moore, 1970). Contact also plays an important role in
creating representations of areas (e.g. Haartsen, 2002). Most of the searchers
mentioned in the interview that they had already had direct experience of their
search area. The local searchers searched within their current residential
environment. The regional searchers also mainly knew their search area from
previous residential experiences or, for example, through recreational activities.
However, the distant searchers also had previous direct or indirect contact with
most of the areas they were searching in. Some of the distant searchers started
their search close to their current residence and knew these areas by living in
their proximity. One couple (D10) partly searched in areas she and her husband
had lived in before. Her husband was the only distant searcher who had
previously lived in the northern Netherlands. Direct experience was sometimes
also as a result of holidays or visits for other reasons, as illustrated by D8 who
130
became acquainted with his later search area in northern Groningen through
having a girlfriend who lived in the city of Groningen:
… so at the weekends I visited her. Sometimes we went for trips in the
surroundings, in that way I got to know Groningen really well, of
course.
Sometimes it was a more a case of indirect contact with the North in general, for
example through having had a nice visit to the city of Groningen (without
visiting the rural surroundings) or having a mother who grew up in the city of
Groningen. D7 described a clear example of how direct contact with an area can
change the representation of an area in a positive way. She did not know the
town of Veendam and its surroundings, while her husband knew the area
through his work and proposed looking for a house there because of a job
opportunity. While at first she had a very negative idea of the region, a visit to
the area changed this drastically, giving her a positive perception:
Gosh, all the way up north, I couldn’t bear thinking about it. Veendam
didn’t sound nice at all. (…) We went to Veendam and I liked the place
very much. And beautiful houses, I said ‘I could really live here’.
The difference between the awareness space and the search space in the
theory about the residential search process implies that knowledge of an area
does not necessarily lead to a positive appreciation of the area. It means that
based on certain perceived characteristics, some areas are excluded from the
search process before it actually starts. On the basis of the interviews, this indeed
appeared to be the case. This was discussed with the local searchers more
hypothetically, because they all intended to continue searching in their current or
neighbouring village. Both local and regional searchers mentioned areas within
the province they did not want to live in, based on indirect or direct contact. The
latter is illustrated by this quote from R5:
131
I worked as a trainee near Erica. Well, I do not exactly like the people
there. The way they speak is completely different. They are blunt. I
was young then, but my first reaction when I got home was: ʹI donʹt
want to have anything to do with them, I wouldnʹt dream of living
there’.
Some of these searchers also mentioned specific villages they wanted to avoid.
Two of the local searchers were even more specific, by saying that they also knew
which street or neighbourhoods to avoid within a village. L1 explains how she
distinguishes between different villages surrounding her current village:
Those villages around Hallum, you‘d want to live in some of them, but
in some of them you wouldn’t want to live at all. Take Blija for
example, that really is a boorish village. The youths go to a ‘keet’ (rural
youth hangout, cf. Haartsen and Strijker, 2010), and that then is their
goal in life. Well, my goals are different. It may have to do with
education. Because what I just said about Blija, those people really
seem more simple‐minded.
The difference with distant movers is that they refer to whole provinces instead
of smaller areas or villages when speaking about areas which are not considered
in the search process based on perceived characteristics. These different levels of
scale used by the different groups of searchers are comparable with the finding
that the reputation of a neighbourhood is more detailed for residents than for
non‐residents (Permentier, 2009). D8 remarks:
Brabant, I wouldnʹt want to live there. Look, of course parts of Brabant
are very beautiful. (…) But I have had some very unpleasant
experiences with people from Brabant. I always say that people from
Brabant are sneaky, they act nicely to your face, but behind your back
they say something entirely different. (…) Somebody from the North
will make it clear that he likes you or doesnʹt like you. Even then it
132
may take some time before he becomes friendly. You may call it
stubborn, but at least it is honest.
This quote again shows that direct contact by knowing people in an area does not
always lead to a more positive representation. Nevertheless, as soon as the
barrier raised by a negative perception of an area is broken, a perception can
change, as shown by another quote from D8:
... and Friesland was a obstacle we couldnʹt actually overcome at first.
Also because of the reputation of the Frisians being a bit stubborn.
And the language, you know. To us westerners it seems that all
Frisians speak Frisian. (…) And then when you get there you donʹt
hear much Frisian being spoken at all. And you go to a friend in
Goingarijp and you see that Friesland is actually really beautiful. And
then you drive past this place and you think this is a place where I
might want to live. Yes, thatʹs how things go.
Representations of areas not only play a role in determining the initial
search area, they can also change or become more refined during the search
process and therefore influence which areas to omit or to focus on in the search
process. The representations of areas that local and regional searchers had did
not change very much, probably because they did not really explore areas that
were new to them. The distant searchers did explore areas that were relatively
unknown to them. When they did, sometimes their impression of the province
they were searching in changed negatively, for example, finding it not as quiet as
expected or disliking the people living there. Sometimes this results in omitting
that area from further searches. Developing search activities can also lead to a
more differentiated representation of a province, when searchers discover areas
within the wider area which inspire different images or feelings. They described
how they narrowed their search area based on these perceptions. D8 recounted
the poor impression the east of the province of Groningen made on him:
133
Well, this is also a region, itʹs really a very beautiful area, you know,
Bourtange, all those places. (...) Only you see straightaway that the
people are not well off, oh yes, you can really see that (...) Yes, the
houses, they simply look a bit shabby. There also isnʹt very much to do
anyway, I think.
On the other hand, the northern part of the province made a very creative
impression on him, which better fit with his and his partner’s lifestyle and job:
And thatʹs something we really pay attention to, for example in
Eenrum. There you see people who own small galleries. And that gave
us the idea, oh, creative people must live here. (....) Those must be
people who came to live here, or already lived here, involved in
making art.
From the answers this distant searcher recorded in his diary questionnaire, it
became clear how his perceptions of villages, which he obtained during when
visiting available properties, influenced the development of a preference for
specific villages within his search area. After visiting a house in Oude Bildtzijl he
noted ‘nice surroundings, nice village, beautiful landscape’, while a visit to a
house in Sijbrandahuis (both villages are in northern Friesland) led him to
remark ‘boring landscape, no facilities nearby, somehow I like the people in
Dokkum and surroundings less than the people in Oude Bildtzijl and
surroundings’. Oude Bildtzijl became one of his preferred villages. However, a
few weeks later he wrote ‘we have visited the North again a couple of times and
we think that the area around Eenrum and Pieterburen is simply great’. Quite
soon after this he bought a house in Pieterburen (in northern Groningen).
The searchers themselves acknowledge that some degree of coincidence is
involved in developing either a positive or a negative perception of an area. This
is illustrated by the following quote from D8:
We went to Wierum, we got there sometime around dinnertime.
Wierum really is a lovely, picturesque village and we ended up in a
134
pub and the landlord fried some fresh fish for us and that was really
very cosy, that definitely has some influence on your judgment.
Notably, when referring to perceived characteristics based on which places
are excluded from the search process beforehand or during the search process,
the respondents often referred to social characteristics. Perhaps this is
unsurprising, given that social characteristics also play an important role in the
neighbourhood reputation concept (Permentier, 2009). However, the focus in
literature on rural migration is most often on the importance of landscape
amenities to determining residential choice (e.g. Argent et al., 2007).
According to Koopman (2012), relying on neighbourhood reputation is one
way of dealing with the uncertainty about the social qualities of an area. This
uncertainty is greatest for the distant searchers because they are searching in
areas less well known to them. It appears that some of them use the reputations
of provinces to narrow down their search space. However, when it comes to
really deciding where to live within this more limited search area, these distant
searchers do not rely on this ‘risk‐minimizing strategy’ because social
characteristics are too important to their choices. Therefore, they search in areas
they know to some extent and they spend a great deal of effort in getting to know
these areas personally. This also becomes clear from their search activities,
discussed in the next section. A preference for a specific housing type was key to
the search of two other distant searchers. Their social preferences were more
general, such as ‘people behaving normally’. Perhaps because of that, they spent
less effort on the social reputations of areas and in getting to know these areas
thoroughly. For them, what mattered most was finding the right house.
5.4.4 Search activities
It is clear from the interviews that the internet, and in particular the Dutch
housing website ‘Funda’, has developed into a central information source for
local, regional and distant searchers. Searchers used the website to find available
properties using search criteria, but the website also allows the user to receive
notifications of new properties that fit with predetermined preferences by e‐mail.
Most searchers used Funda or comparable websites to find available properties
135
within their defined search space. Nevertheless, some of the distant searchers
also used them to explore new potential search areas. D10 in particular examined
several regions within the Netherlands using the internet. Based on housing
characteristics and prices, she omitted some of these areas from her search space
without visiting them. D9 describes how Funda led him to a new search area.
While his main search area was in southeastern Groningen, his search criteria
apparently also included vacancies in the North of the province. As a result, he
came across a house that appealed to him greatly in the village of Pieterburen. He
and his partner inspected the house, and though they decided not to take it, they
discovered in the meantime that they liked the village and surrounding area very
much. After a time they decided that this would become their main search space.
However, most searchers also employed other search methods as well as
using the internet. The methods used slightly differed among the different
groups of searchers. While several searchers received notifications from a real
estate agent when suitable vacancies became available, two of the local searchers
and a regional searcher mentioned some more personal contact with a real estate
agent in their current or neighbouring village. This kind of contact probably gave
these searchers an advantage in the sense of being notified at an early stage that a
property was becoming available, but this is not explicitly mentioned. Other
‘local’ search methods include being notified when a local housing corporation
sold some of their houses. L3 also indicated that people from the village
sometimes phoned them when they saw a suitable house for sale. R5 used to
cycle or drive around his search area to look for houses he liked, regardless of
whether or not they were for sale. He had some ‘favourites’ which he kept an eye
on until they came onto the market. Two of the local searchers also mentioned
that they sometimes walked or cycled around the village to see whether there
were suitable properties. However, despite the greater distance they had to
cover, driving around looking for properties is something also done by the
distant searchers. Some explained how, when visiting properties, they explored
their search area more thoroughly, also looking for other houses for sale. Two
other distant searchers illustrate how the residential search process is always
present, one couple looking for a house they liked (D10), the other for a ‘good
feeling’ (D8):
136
…and even when we are driving somewhere, our eyes are looking left
and right ‘is there a nice house for sale here?’
Well, then we started making trips all around the Netherlands. At a
certain moment we started putting circles on the map around places
that felt good. Or if I happened to be in a certain area and I thought:
hey, this feels nice, you know, then you put a circle around it.
This idea of ‘it felt good’ is mentioned by other searchers as well, referring to
areas and houses. Inspections appear to be valuable not only for investigating
more deeply those objective characteristics which are not always visible on the
internet, such as the quality of the house, the view from the house, the amount of
facilities in the village, but also less tangible characteristics, as expressed in the
following quote (D8):
Because we feel we really have to fall in love with a house (...). So
perhaps it is more a matter of the heart than the head.
As Koopman (2012) argues, as a non‐resident it is possible to gain insight
into the observable characteristics of an area and also less tangible aspects such as
atmosphere, but the social quality of an area is more difficult to discover. That
residents have an information advantage in this respect is acknowledged in the
interviews with the various searcher groups. The local searchers were aware that
they were better able to differentiate within the area or village than outsiders.
The regional and distant searchers mentioned different strategies for gaining
greater insight into the social qualities of an area. D8 attached great importance
to driving around, talking to people and looking for signs of the presence of
artistic or creative people. R6 mentioned asking people who live in the area about
their perceptions of regions or villages, he also saw the internet as a way of
exploring what kind of activities go on in a village, at schools, etc. Another
example is offered by D9:
137
By just walking around, talking to people. Asking them what it is like
to live there (…). Also, talk to the neighbours when you visit a house.
Ask them how they feel about living there and if there are any
problems. It isnʹt watertight, but you try to minimize the risks this
way.
5.4.5 Resources, restrictions and opportunities in the housing market
While representations can limit the search process in some ways, there can also
be more concrete factors. Obviously, the interaction between the financial
resources of a searcher and what is available in the housing market determines to
a large extent whether and where a move is made. Resources and opportunities
were discussed in almost all the interviews. For the local and most of the regional
searchers, house prices influenced choosing specific houses, but did not influence
the search area. House prices did influence the search area choice of three of the
distant searchers. In the Netherlands the Randstad area generally has higher
house prices than the more peripheral areas (De Groot et al., 2011a). Within the
northern Netherlands the most northern and eastern municipalities have the
lowest house prices in the country (Bijker et al., in press). The distant searchers
explained during the interviews that they discovered in their search processes
that their preferred housing types were practically unaffordable for them in the
central part of the country, which forced them to search elsewhere. Some areas
were then also dropped from the search space due to overly high house prices.
The relatively lower house prices in the northern and eastern parts of the
northern Netherlands played a considerable role in their choice to search there.
D8 describes how he traded‐off between a preference for a certain landscape and
a preference for a certain type of house due to differences in house prices within
the northern Netherlands:
I prefer the woods, but living near the woods is very expensive. (…) If
we consider our budget, that would imply that we could buy a small
terraced house in Dwingeloo and then to be able to enjoy nature we
would have to walk or cycle a few kilometres to get there. It is a
possibility, but we actually really prefer a detached house. Look, I
138
would really be happy to be able to just walk out my door and stand
on my own piece of land.
In addition to financial resources, some other types of restriction came to the fore
in the interviews. First, most distant searchers mentioned social ties or work‐
related ties to the current residential environment. As this quote of D10
illustrates:
...on the one hand there is living in the country, our desire, a nice old
house, but on the other hand there is the large distance from our
family and friends. And my husband has retired, so how are we going
to fill our time? Are we going to do voluntary work or will we still be
able to have the opportunity to do something in our own line of work?
So we really feel that we have to let many things go that are dear to us
and made us very happy.
These ties resemble the local ties of the local searchers, discussed in Section 5.4.2.
However, for the local searchers these ties were so strong that they prevented
them from moving to another place altogether, while for the distant searchers
these ties were more a constraining factor with a possible influence on their
search space. Other restrictions mentioned by distant searchers are the lack of a
car and a driver’s license and municipal building rules. A factor mentioned by
several regional and distant searchers which did not so much restrict the search
space as restrict making a move at all is the possibility of selling the current
house at an acceptable price, which was no longer a self‐evident truth in the
Dutch housing market at that time.
5.5 Conclusion and discussion
This paper investigates how people search for a new place to live in rural areas in
the northern Netherlands using a diary approach, a new method for studying the
search process. Our findings show that the search areas of local, regional and
distant searchers differ in scale and in the extent to which they change during the
139
search process. While local movers spoke about villages, regional and distant
movers initially spoke about areas. Whereas the local movers in our study started
searching in their own or in a neighbouring village and continued to do so
throughout the search process, the regional and distant searchers all experienced
changes in their search area.
This corresponds with the finding of De Groot et al. (2012) that intended
non‐local movers are less likely to realize their rural location preferences than
intended local movers. While their longitudinal approach offers no explanation
for this difference, our findings show that the importance of local ties for local
searchers is an important reason for this distinct difference in search patterns.
Their less strong preference for a specific rural location makes it easier for
regional and distant searchers to change their search area. While local ties keep
the local searchers close to their current place of residence, the same kind of ties
act as a restriction in distant searchers’ search processes. Financial restrictions
often directed the search spaces of the distant searchers to the North, with its
lower house prices, after other rural areas had been excluded from their search
space due to overly high prices (see also Bijker et al., in press). For local and
regional searchers, financial restrictions play more of a role in the choice of
specific houses than in the choice of search areas. Finally, our results suggest that
the better knowledge of the local housing market enjoyed by local movers does
not make it easier for them to realize their preferred move. All searchers use the
internet to gain information about available properties and therefore have
comparable levels of knowledge in this respect. What differs is their knowledge
of the social qualities of an area. Local and regional searchers are able to
distinguish between different villages and even streets. Distant searchers have
developed several strategies to overcome this disadvantage, ranging from
driving around to meeting the neighbours when visiting a house. The importance
that is attached to detecting these social qualities, and also the importance of
certain feelings about areas and houses, makes the internet has not replaced
‘older’ search methods such as driving around and visiting houses. However, it
appears that internet sites such as Funda make it easier for distant searchers to
explore new potential search areas, sometimes also resulting in changes in their
search spaces.
140
Representations of areas play an important role in the search process in
different ways. First, direct or indirect contact with an area appears to be
essential to include the area in the search space. However, knowledge of an area
does not necessarily lead to a positive appreciation of the area. Both local and
regional searchers mentioned areas within the province or specific villages in
which they did not want to live, and distant movers referred to whole provinces
which were omitted based on perceived characteristics. However, in the process
of developing their search activities, searchers discovered areas that inspired
different images or feelings within the wider area. Of note is that when referring
to perceived characteristics based on which places were excluded from the search
process, the respondents often referred to social characteristics. In a rural context
the focus is often on the importance of landscape and locational amenities in
determining residential choice (e.g. Argent et al., 2007). Our findings suggest that
more attention is needed for the perceived social characteristics of rural areas,
which has already been acknowledged in an urban context (e.g. Permentier,
2009).
Using a diary approach provides more insight into the residential search
process, by revealing the changes in the search space and preferences that are
made between the beginning and the end of the search. Following searchers over
time shows the non‐linearity of the process and the role played by
representations. These aspects are incorporated in residential choice models such
as that of Brown and Moore (1970), but remain invisible when using longitudinal
methods with longer intervals. The qualitative elements elucidate the more
emotional, less tangible aspects of residential decision‐making (see also Levy et
al., 2008) and the influence of coincidence on the process. These aspects are also
overlooked when using a large quantitative longitudinal dataset. The
combination of an in‐depth interview with the questionnaire turned out to be
very useful. The personal contact and the fact that the researcher made the effort
to come and visit the respondents led to high compliance with respect to
completing the questionnaire. The interview provided rich information, which
also helped interpret the questionnaire answers. Due to its more quantitative
nature, the questionnaire provided most information on the development of the
141
search process. The final decision, involving trade‐offs between preferences, is
harder to disentangle from it. One way of solving this would be to add a ‘final
interview’ to the approach. The design of the questionnaire could be
reconsidered on some points. For instance, the finding that half of the
respondents did not change their preferences could be influenced by the fact that
we showed the respondents what they filled in the last time. After some time
completing the questionnaire, ‘habituation’ (Bolger et al., 2003) starts and it is
perhaps easy to just move on to the next question. A point of concern is the
recruitment of participants. We tried to find searchers through real estate agents,
which was not very easy, as explained in the methodology part. An idea would
be to try and approach respondents somehow through housing websites on the
internet. With regard to the diary approach technology, it might be worthwhile to
explore the possibilities of the increased use of smartphones, which permit
people to note down their experiences during the search process in real time, for
instance when visiting houses.
Acknowledgements
The authors would like to thank all the respondents for their time and effort,
which made this research possible. Further, we would like to thank the following
real estate agents for sharing information and their help in getting in touch with
house seekers: Annemiek de Bie, Annen; Bruintjes en Keurentjes, Winsum; De
Waard Makelaars, Franeker; Germeraad en Germeraad, Stiens; G2 Makelaars,
Franeker; Harm Stuut Makelaardij, Heiligerlee; Hup en Fidom, Hoogeveen; Ing.
H.H. Smit Makelaardij, Zuidbroek; Koomans‐Venema Makelaardij, Oude Pekela;
Makelaardij Koomans, Gasselternijveen; Oosterveld Makelaardij, Onstwedde;
Regiomakelaardij, Dokkum; Rita Sterkenburg, Ulrum; Roelof Berg Makelaardij,
Diever; Schelhaas Makelaardij, Hoogeveen; Tjerk Dijkstra Makelaardij, Dokkum;
Varwijk & Sibma, Ureterp; Viva Makelaars, Winschoten.
References
Andersen, H. S. (2011) Explanations for long‐distance counter‐urban migration
into fringe areas in Denmark, Population, Space and Place, 17, pp. 627‐641.
142
Argent, N., Smailes, P. & Griffin, T. (2007) The amenity complex: towards a
framework for analysing and predicting the emergence of a
multifunctional countryside in Australia, Geographical Research, 45, pp. 217‐
232.
Barrett, F. (1976) The search process in residential relocation. Environment and
Behavior, 8(2), pp. 169‐198.
Bijker, R. A. & Haartsen, T. (2012) More than counter‐urbanisation: migration to
popular and less‐popular rural areas in the Netherlands, Population, Space
and Place, 18, pp. 643‐657.
Bijker, R. A., Haartsen, T. & Strijker, D. (in press) Migration to less‐popular rural
areas in the Netherlands: exploring the motivations, Journal of Rural
Studies, doi: 10.1016/j.jrurstud.2012.07.003.
Bolger, N., Davis, A. & Rafaeli, E. (2003) Diary methods: capturing life as it is
lived. Annual Reviews of Psychology, 54, pp. 579‐616.
Boyle, P. & Halfacree, K. (1998) Migration into Rural Areas: Theories and Issues
(Chichester: John Wiley & Sons).
Brown, L. A. & Moore, E. G. (1970) The intra‐urban migration process: a
perspective. Geografiska Annaler, 52(1), pp. 1‐13.
Bunce, M. (1994) The Countryside Ideal. Anglo‐American Images of Landscape
(London: Routledge).
Clark, W. A. V. & Flowerdew, R. (1982) A review of search models and their
application to search in the housing market, in: W.A.V. Clark (Ed.)
Modelling Housing Market Search, pp. 4‐29 (London: Croom Helm).
Cronin, F.J. (1982) The efficiency of housing search, Southern Economic Journal, 48,
pp. 1016‐1030.
De Groot, C., Mulder, C. H. & Manting, D. (2011a) Intentions to move and actual
moving behaviour in the Netherlands, Housing Studies, 26, pp. 307‐328.
De Groot, C., Mulder, C. H., Das, M. & Manting, D. (2011b) Life events and the
gap between intention to move and actual mobility, Environment and
Planning A, 43, pp. 48‐66.
De Groot, C., Daalhuizen, F. B. C., Van Dam, F. & Mulder, C. H. (2012) Once an
outsider, always an outsider? The accessibility of the Dutch rural housing
143
market among locals and non‐locals, Journal of Rural Studies, doi:
10.1016/j.jrurstud.2011.12.001.
Donaldson, B., (1973) An empirical investigation into the concept of sectoral bias
in the mental maps, search spaces and migration patterns of intra‐urban
migrants, Geografiska Annaler, 55(1), pp. 13‐33.
Gkartzios, M. & Scott, M. (2009) Residential mobilities and house building in
rural Ireland: evidence from three case studies, Sociologia Ruralis, 50, pp.
64‐84.
Goetgeluk, R. (1997) Bomen over Wonen: Woningmarktonderzoek met
Beslissingsbomen (Utrecht: Koninklijk Nederlands Aardrijkskundig
Genootschap/Universiteit Utrecht).
Grimsrud, G. M. (2011) How well does the ‘counter‐urbanisation story’ travel to
other countries? The case of Norway, Population, Space and Place, 17, pp.
642‐655.
Haartsen, T. (2002) Platteland: Boerenland, Natuurterrein of Beleidsveld? Een
Onderzoek naar Veranderingen in Functies, Eigendom en Representaties van het
Nederlandse Platteland (Utrecht/Groningen: Koninklijk Nederlands
Aardrijkskundig Genootschap/Rijksuniversiteit Groningen).
Haartsen, T., Huigen P. P. P. & Groote P. (2003) Rural areas in the Netherlands,
Tijdschrift voor Economische en Sociale Geografie, 94, pp. 129‐136.
Haartsen, T. & Strijker, D. (2010) Rural youth culture: Keten in the Netherlands,
Journal of Rural Studies, 26, pp. 163‐172.
Haartsen, T. & Venhorst, V. (2010) Planning for decline: anticipating on
population decline in the Netherlands, Tijdschrift voor Economische en
Sociale Geografie, 101, pp. 218‐227.
Halfacree, K. (1994) The importance of ‘the rural’ in the constitution of
counterurbanization: evidence from England in the 1980s, Sociologia
Ruralis, 34(2‐3), pp. 164‐189.
Harper, S. (1991) People moving to the countryside: case studies of decision‐
making, in: T. Champion & C. Watkins (Eds) People in the Countryside:
Studies of Social Change in Rural Britain, pp. 22‐37 (London: Paul Chapman
Publishing).
144
Holloway, L. & Hubbard, P. (2001) People and Place. The Extraordinary Geographies
of Everyday Life (Dorchester: Dorset Press).
Hooijmeijer, P. & Oskamp, A. (1996) A simulation model of residential mobility
and housing choice, Netherlands Journal of Housing and the Built
Environment, 11(3), pp. 313‐336.
Hortulanus, R. P. (1995) Stadsbuurten: Bewoners en Beheerders in Buurten met
Uiteenlopende Reputaties (Den Haag: VUGA).
Huff, J. O. (1982) Spatial aspects of residential search, in: W. A. V. Clark (Ed.)
Modelling Housing Market Search, pp. 106‐129 (London: Croom Helm).
Huff, J. O. (1986) Geographic regularities in residential search behaviour, Annals
of the Association of American Geographers, 76(2), pp. 208‐227.
Kan, K. (1999) Expected and unexpected residential mobility, Journal of Urban
Economics, 45, pp. 72‐96.
Koopman, M. (2012) Economic Analysis of Neighbourhood Quality, Neighbourhood
Reputation and the Housing Market (Amsterdam: IOS Press).
Larson, R. (1989) Beeping children and adolescents: a method for studying time
use and daily experience, Journal of Youth and Adolescence, 18(6), pp. 511‐
530.
Laurenceau, J. & Bolger, N. (2005) Using diary methods to study marital and
family processes, Journal of Family Psychology, 19(1), pp. 86‐97.
Levy, D., Murphy, L. & Lee, C. K. C. (2008) Influences and emotions: Exploring
family decision‐making processes when buying a house, Housing Studies,
23, pp. 271‐289.
Lewis, G. J. (1998) Rural migration and demographic change, in B. Ilbery (Ed.)
The Geography of Rural Change, pp. 131‐160 (Harlow: Pearson Education
Limited).
Litt, M. D., Cooney, N. L. & Morse, P. (1998) Ecological momentary assessment
(EMA) with treated alcoholics: methodological problems and potential
solutions, Health Psychology, 17, pp. 48‐52.
McPeake, J. (1998) Religion and residential search behaviour in the Belfast urban
area, Housing Studies, 13(4), pp. 527‐548.
Moscovici, S. (1982) The coming era of representations, in J. P. Codol & J. P.
Leyens (Eds) Cognitive analysis of social behavior: proceedings of the NATO
145
Advanced Study Institute on ‘The Cognitive analysis of social behavior’, pp. 9‐45
(The Hague: Nijhoff).
Mulder, C. H. (1993) Migration: A Life Course Approach (Amsterdam: Thesis).
Mulder, C. H. (1996) Housing choice: assumptions and approaches, Netherlands
Journal of Housing and the Built Environment, 11(3), pp. 209‐232.
Niedomysl, T. (2010) Towards a conceptual framework of place attractiveness: A
migration perspective, Geografiska Annaler: Series B: Human Geography, 92,
pp. 97‐109.
Palm, R. & Danis, M. (2002) The internet and home purchase, Tijdschrift voor
Economische en Sociale Geografie, 93, pp. 537‐547.
Permentier, M. (2009) Reputation, Neighbourhoods and Behaviour (PhD thesis
Faculty of Geography, University of Utrecht).
Phipps, A. G. (1983) Utility function switching during residential search,
Geografiska Annaler, 65(1), pp. 23‐38.
Richardson, A. (1994) The health diary: an examination of its use as a data
collection method, Journal of Advanced Nursing, 19, pp. 782‐791.
Rieman, J. (1993) The diary study: a workplace‐oriented research tool to guide
laboratory efforts, Conference on Human Factors in Computing Systems.
Proceedings of the INTERACT ʹ93 and CHI ʹ93 conference on Human factors in
computing systems, pp. 321‐326.
Rossi, P. H. (1955) Why families move (Glencoe, IL: Free Press).
Short, J. R. (1991) Imagined Country. Society, Culture and Environment (London:
Routledge).
Speare, A. (1974) Residential satisfaction as an intervening variable in residential
mobility, Demography, 11, pp. 173‐188.
Stockdale, A. (2006) Migration: Pre‐requisite for rural economic regeneration?
Journal of Rural Studies, 22, pp. 354‐366.
Stone, A. A., Kessler, R. C. & Haythornthwaite, J. A. (1991) Measuring daily
events and experiences: decisions for the researcher, Journal of Personality,
59(3), pp. 575‐607.
Sudman, S. & Ferber, R. (1971) Experiments in obtaining consumer expenditures
by diary methods, Journal of the American Statistical Association, 66(336), pp.
725‐735.
146
Van Dam, F., Heins, S. & Elbersen, B. S. (2002) Lay discourses of the rural and
stated and revealed preferences for rural living: some evidence of the
existence of a rural idyll in the Netherlands, Journal of Rural Studies, 18, pp.
461‐476.
Voets, S. (1994) The Concept of Choice and Relocation Behaviour (Amsterdam:
PDOD‐UvA).
Walmsley, D. J., Epps, W. R. & Duncan, C. J. (1998) Migration to the New South
Wales North Coast 1986‐1991: Lifestyle motivated counterurbanisation,
Geoforum, 29(1), pp. 105‐118.
Walmsley, D. J. & Lewis, G. J. (1993) People and Environment: Behavioural
Approaches in Human Geography (Harlow: Longman Scientific & Technical).
147
6. Conclusions
This dissertation contributes to the critical reconsideration of the concept of
counterurbanisation by studying migration to less‐popular rural areas. We have
examined the characteristics and motivations of movers to these less‐popular
rural areas. To determine the extent to which the results are specific for less‐
popular areas, a direct comparison with in‐migrants to popular areas in the same
region was included in the study. The results are also compared with existing
outcomes and ideas in the literature on counterurbanisation. In addition to these
questions which focus on the outcome of the residential mobility process, we
gained greater insight into the residential search process before a move. We used
a combination of different data and methods to answer our research questions.
We studied the characteristics and motivations of in‐migrants by collecting our
own survey data, as well as making use of secondary data. To investigate the
residential search process in a rural context, we conducted a small‐scale study
using a diary approach, a new method for studying the residential search
process. This diary approach consisted of in‐depth interviews followed by an
electronic questionnaire which was sent to the respondents every two weeks.
6.1 Less‐popular and popular rural areas in the northern Netherlands
The popularity of rural areas as places to live is defined in this dissertation using
average house price per municipality. House prices are an indication of what
people are willing to pay for houses and their surroundings (Luttik, 2000; Visser
and Van Dam, 2006), thus reflecting the value buyers attach to them. This
average house price incorporates the residential environment as well as housing
characteristics such as size and quality. The use of house prices instead of in‐
migration numbers, which are often used in other studies of rural areas’
popularity (Argent et al., 2007; McGranahan, 2008), is more appropriate to the
Dutch context. In the Netherlands, population growth at the municipal level is
strongly influenced by government development policy (Marlet, 2009). New
housing is only allowed in a limited number of areas, while in others building
148
houses is very restricted, especially in rural areas, both in the open countryside
and in villages (Van Dam et al., 2002). These restrictions make it impossible for
the Dutch housing market to react to changes in demand for new houses, which
by increasing demand leads to price increases rather than increased in‐migration
(Marlet, 2009). Rental prices differ little between regions in the Netherlands,
which is why these are not considered in defining the popularity of rural areas
for this study.
Based on house prices, we distinguished three types of rural areas in the
northern Netherlands: less‐popular rural areas, average rural areas and popular
rural areas (see also Chapter 2). These areas are considered rural based on
Statistics Netherlands’s national standard for the degree of urbanisation, the so‐
called address density, consisting of municipalities with on average fewer than
1,000 addresses per square kilometre. In this dissertation we focus on less‐
popular areas, but we also compare with popular rural areas. The less‐popular
areas are located along the northern and eastern borders of the northern
Netherlands. They can be characterised as having an open, marine clay landscape
with a large share of large‐scale arable farmland. This is consistent with the
findings of Ulrich (1986) and McGranahan (2008), showing that people prefer
areas with a mix of forest and open land and relatively little cropland. Compared
to the popular areas, they have relatively fewer natural and recreational areas
and they have a relatively low share of hotel and catering industry employment.
The latter was also found to be an indicator of the popularity of rural areas by
Argent et al. (2007). Other studies found accessibility to be an important
determining factor for the popularity of rural areas (e.g. Woods, 2005). The less‐
popular areas in our study have a comparable travel distance to urban centres in
the North, but are more distant from the centre of the Netherlands than popular
areas. It seems that in this case, location relative to the centre of the country is
more important than location within the northern Netherlands. The people who
inhabit these areas have a lower average household income than those in the
popular rural areas. In these less‐popular areas, both population decline and a
decline in the number of households are expected, and in some municipalities
population decline is already taking place (Haartsen and Venhorst, 2010).
Popular rural areas, in contrast, can be characterised as having a more semi‐open
149
landscape with predominantly grazing livestock farming, a landscape type that is
generally perceived as attractive (McGranahan, 2008; Ulrich, 1986). They also
have a relatively large share of natural areas. Popular rural areas have an
expected household growth and are less distant from the centre of the
Netherlands.
Looking at the northern Netherlands as a whole, a region in which 10% of
the Dutch population (1,700,000) lives, the net migration is slightly negative. A
five‐year annual average of 22,326 people (2003‐2007) left the North for elsewhere
in the Netherlands, while 21,913 arrived. Of these in‐migrants from elsewhere in
the Netherlands, 54% found a place to live in one of the 58 municipalities
classified as rural based on address density. Looking at migration within the
region in the same period, 31,374 people (61% of the total group of inter‐
municipal migrants within the North) moved to a rural municipality from either
an urban or rural municipality within the northern Netherlands. Regarding the
three types of areas, the less‐popular, average and popular rural areas in the
northern Netherlands all show a negative net migration for the period 2003–2007,
and the size of the net migration rate does not differ significantly between the
three types of areas. This means that more people left an area than moved to it,
even for the popular areas. The average number of in‐migrants per 1,000
inhabitants is around 40 for the three types of areas in the same period. Thus,
popular rural areas do not attract significantly more migrants than less‐popular
rural areas. There is a considerable migration flow between urban areas and
popular rural areas, while the migration flow between less‐popular and average
rural areas on the one hand, and popular rural areas on the other, is very limited.
6.2 Who moves into less‐popular areas and why?
Our study of migration to less‐popular rural areas in the northern Netherlands
(Chapter 3) based on newly collected survey data, shows that the migrants to
these areas are more diverse than the middle‐class movers often associated with
migration to the rural (see e.g. Halfacree, 2008). However, the migrants in our
study are also more mixed than the low‐income groups found moving to
impoverished rural areas in the US (Fitchen, 1995; Foulkes and Newbold, 2008)
150
and the movers to depopulating areas in Scotland (Stockdale, 2006). Although
they have a relatively low income in the Dutch context, the lowest‐income group
is only small. Combined with the relatively large proportion of working people,
the variety of sectors they work in, and the large group with higher education
qualifications, our movers seem to correspond better to the more diverse group
of movers to fringe areas in Denmark as found by Andersen (2011). This group of
movers includes quite a large proportion of migrants on low incomes, but also
encompasses movers with middle or high incomes trying to realise housing
preferences they could not afford in more expensive areas. In addition to these
socioeconomic characteristics our study found that the movers to less‐popular
areas are predominantly young, while a relatively small proportion moved from
urban areas. The mobility circles vary from local – from surrounding
municipalities – to longer distance – from elsewhere in the Netherlands, but the
majority made a regional move from within the northern Netherlands.
For migration into rural areas in general, the quality of the residential
environment is often seen as the most important motivation (e.g. Gkartzios and
Scott, 2009, Halfacree, 2008). For movers to less‐popular rural areas, it seems that
other motivations such as marriage, the proximity of family and friends,
employment considerations and housing reasons are more important (Fitchen,
1995; Foulkes and Newbold, 2008; Grimsrud, 2011; Stockdale, 2006). However,
based on existing studies it was unclear to what extent the typical ‘quality‐of‐life’
motivations connected to the counterurbanisation concept hold true for migrants
to less‐popular rural areas (Foulkes and Newbold, 2008; Stockdale, 2006). Our
results show that these quality‐of‐life motivations do play a role in decisions to
move to less‐popular rural areas, which corresponds to the findings of Foulkes
and Newbold (2008). However, disentangling them into motivations for moving
to a rural area, in general, and to the specific rural area, in particular, has proved
that these quality‐of‐life motivations are especially important in the choice for
rural living in general. Housing characteristics were the main motivation for
moving to the specific rural area, directly followed by the physical qualities of the
environment and personal reasons, such as moving in with a partner and living
close to family and friends. Low house prices in the area only came fifth in the
hierarchy of motivations.
151
Combining the characteristics and motivations of the movers reveals the
diversity within the group of movers to less‐popular rural areas even more (see
Chapter 3). The physical qualities of the environment attract a group of higher
educated movers with high incomes, and middle‐aged people, a group very
similar to the stereotypical counter‐urbanite (e.g. Milbourne, 2007). Furthermore,
our analysis reveals a group of movers motivated by the proximity to family and
friends, consisting of return migrants, singles, the youngest and oldest age
groups and also the lowest‐income group. Low house prices are an important
motivation for young movers, single people and for people moving from
elsewhere in the Netherlands, but not for the low‐income migrants. Work‐related
motivations are important for highly educated movers, people moving from
surrounding municipalities and the 35‐44 age group. The housing characteristics
motive usually refers to the availability of a specific house and is mentioned by a
diverse group of movers, but in particular by people moving from urban areas.
6.3 Different areas, different people?
To determine to what extent these results are really specific for less‐popular rural
areas, we also made a direct comparison with the characteristics and motivations
of in‐migrants to popular rural areas in the same region. With respect to
migration statistics, in Chapter 2 it appears that less‐popular areas attract a larger
share of movers from within the northern Netherlands, a larger inflow from rural
areas and generally a younger group of movers. The multivariate analysis of
secondary survey data in the same chapter showed that people who moved to
their current address motivated by a desire to live closer to family and friends
were more likely to have moved to less‐popular areas. Highly educated people
and people moving from urban areas were more likely to have made a move to
popular rural areas. Based on this, we concluded that migration patterns in
different types of rural areas within countries cannot sufficiently be explained by
the classical conceptualisation of counter‐urbanisation as the movement of
middle‐class people from the city in search of a new life in an idyllic rural setting
(see Halfacree, 2008). The characteristics of the movers to popular rural areas do
fit very well within the counter‐urbanisation story, while less‐popular rural areas
152
share personal reasons as an important motive for in‐migration with more
remote rural areas in Europe (see Grimsrud, 2011).
We further investigated the question of whether different types of rural
areas within countries attract different migrants with different motivations in
Chapter 4. Here, we used the survey data we collected ourselves, which
permitted a more comprehensive analysis. In the analysis in Chapter 2
motivations for location choice were mixed with motives for leaving the previous
residence, despite the fact that the reasons for leaving a place can differ from the
motives for choosing a destination (see also Bolton and Chalkley, 1990). In the
follow‐up analysis using our own survey data we were able to focus specifically
on motivations for choosing the destination area. We were also able to include
return migration in our analysis and a more detailed measurement of the
previous residence of the movers. In addition to characterising the movers using
‘traditional’ sociodemographic characteristics, we also used people’s values.
When directly comparing the movers to popular and less‐popular areas in
this logistic regression analysis in Chapter 4, it appears that movers to popular
areas more often belonged to higher‐income groups and were more often more
highly educated. They had also more often previously lived in the municipality
and more often belonged to older age cohorts. The physical qualities of the
environment (i.e. nature, space and the presence of water), work‐related
considerations and the location of the area were more often mentioned by them
as motivations to move to an area. When looking at the eight values included in
the analysis, it appears that compared to movers to less‐popular areas, they
attached more importance to pleasure in life (with the addition ‘enjoyment,
satisfaction’).
The analysis shows that movers to less popular areas were more likely to
have moved from elsewhere in the northern Netherlands than movers to popular
areas. While the study of Stockdale (2006) suggested that less‐popular rural areas
mainly attract movers over short distances, our multivariate analysis showed that
popular and less‐popular areas attract movers from elsewhere in the Netherlands
to the same extent. Movers to less‐popular areas are more likely to belong to the
youngest age group. Accordingly, less‐popular areas attract a young group of
movers, also when directly compared to movers to popular areas (cf. Chapter 3).
153
Compared to the movers to popular areas they more often mentioned moving in
with their partners and the low house prices in an area as motivations for moving
there. The importance of these two motivations corresponds with the findings of
studies focused specifically on less‐popular areas (Fitchen, 1995; Foulkes and
Newbold, 2008; Stockdale, 2006). Movers to less‐popular areas mention the
physical qualities of the environment less often, in contrast mentioning social
qualities such as ‘freedom’, ‘friendliness’ and ‘nice atmosphere’ more often. The
other motivations included in the model do not have a significant effect on
moving to a popular or a less‐popular area. Based on the findings in Chapter 2,
which corresponded to the outcomes in other studies (Fitchen, 1995; Foulkes and
Newbold, 2008; Grimsrud, 2011), we expected that living close to family and
friends would be a more important motivation for movers to less‐popular areas.
However, when including only the motivations for choosing the destination area
in the analysis, instead of mixing the motivations for leaving the previous
residence with those for choosing the destination, it appears that the importance
of this motivation does not differ for movers to popular and less‐popular areas.
When looking at the values included in the analysis, we see that compared to
movers to popular areas, movers to less‐popular areas attach more importance to
being rational in their lives (with the addition in the questionnaire of ‘intelligent,
logical, thoughtful’). They also attach more importance to variety in their life.
This value refers to ‘a life full of challenge, novelty and change’.
Halfacree (2008, 2012) has developed a more general and flexible model for
counterurbanisation, which focuses on the motivations of the rural in‐migrants.
He distinguishes a group labelled ‘mainstream counterurbanisation’, for whom
the rural environment is important but balanced by more practical
considerations, such as proximity to employment or services. The group labelled
‘default counterurbanisation’ is a group for which the rural character of the place
is almost totally incidental and for whom instrumental considerations – for
example relating to employment or family support – are more important
(Halfacree, 2008). Based on studies focusing on one type of rural area, migration
to popular rural areas is often associated with ‘mainstream counterurbanisation’,
while migration to less‐popular areas is often seen as ‘default
counterurbanisation’.
154
The results of our analysis in Chapter 4 show that when using a more
comprehensive multivariate analysis to compare movers to popular and less‐
popular areas, the differences between the two groups are less distinct than in the
analysis in Chapter 2, and less distinct than indicated here. Although our current
analysis does show that popular rural areas attract middle‐class movers more
often, it appears that both types of area attract movers from urban areas. When
considering motivations, movers to less‐popular areas are more often motivated
by low house prices and the personal motivation of moving in with a partner,
which suggests that moving to less‐popular areas can indeed be defined as
‘default counterurbanisation’. However, the finding that movers to popular areas
are more often motivated by instrumental considerations such as work‐related
reasons and the location of the area than movers to less‐popular areas, and that
the motivation to live close to family and friends is of the same importance to
movers to either type of area, suggest that popular areas also attract a group of
‘default counterurbanisers’. Moreover, when identifying the physical and social
aspects of the rural character of the environment, which is the main motivation
for the category of ‘mainstream counterurbanisation’, it appears that while the
physical aspects are important for moving to popular areas, social aspects are
more important for moving to less‐popular areas. Apparently, the rural character
of the destination also matters for those moving to less‐popular areas.
6.4 The residential search process in rural areas
In Chapter 5 we investigated how people search for a new place to live in rural
areas in the northern Netherlands using a diary approach, a new method to
study the search process. Our findings show that the search areas of local,
regional and distant searchers differ in scale and in the extent to which they
change during the search process. While local movers speak about villages,
regional and distant movers initially speak about areas. Whereas the local movers
in our study started searching in their own or a neighbouring village and
continued doing so throughout the search process, the regional and distant
searchers all experience changes in their search areas.
155
This corresponds with the finding of De Groot et al. (2012) that intended
non‐local movers are less likely to realise their rural location preferences than
intended local movers. While their longitudinal approach offers no explanation
for this difference, our findings show that the importance of local ties for local
searchers is an important reason for this distinct difference in search patterns. For
regional and distant searchers, their weaker preference for a specific rural
location makes it easier to change search area. We also found that while local ties
keep the local searchers close to their current place of residence, the same kind of
ties act as a restriction in the search process of distant searchers. Financial
restrictions often directed the search space of the distant searchers to the North,
with its lower house prices, after other rural areas had been dropped from their
search space due to high prices (see also Chapter 3). For local and regional
searchers, financial restrictions play more of a role in the choice for specific
houses than in the choice of search areas. Finally, our results suggest that it is not
the greater knowledge of the local housing market that makes it easier for local
movers to realise their preferred move. All searchers make use of the internet to
gain information about buying opportunities and they therefore have comparable
levels of knowledge of availability. What differs is their knowledge of the social
qualities of an area. Local and regional searchers are able to distinguish between
different villages and even streets. Distant searchers developed several strategies
to overcome this disadvantage, ranging from driving around, to meeting the
neighbours when visiting a house. The importance attached to detecting these
social qualities and also the importance of certain feelings about areas and houses
makes the internet has not replaced ‘older’ search methods such as driving
around and visiting houses. However, it appears that real estate websites such as
Funda in the Netherlands make it easier to explore new potential search areas for
distant searchers, sometimes also resulting in a change of their search space.
Representations of areas play an important role in the search process in
different ways. Firstly, direct or indirect contact with an area appears to be
essential for including the area in the search space. However, knowledge of an
area does not necessarily lead to a positive view of the area. Both local and
regional searchers mentioned areas within the province or specific villages that
they did not want to live in, and distant movers referred to whole provinces to be
156
omitted, based on perceived characteristics. However, when developing search
activities, searchers discovered areas within the wider area that invoked different
images or feelings. Of note is that when referring to perceived characteristics
based on which places were excluded from the search process, the respondents
often referred to social characteristics.
6.5 Counterurbanisation reconsidered
Our results empirically support the idea that counterurbanisation is indeed a
‘complex and differentiated phenomenon’, as argued by Woods (2011: p. 184).
This is true when looking at migration to less‐popular rural areas (see Chapter 3),
but also when comparing migration flows to different types of rural areas (see
Chapter 4). With regard to the less‐popular areas, the analysis we performed that
linked motivations to the characteristics of movers was particularly helpful in
revealing the diversity within the movers group. We have seen that less‐popular
areas also attract middle‐class movers motivated by aspects of the rural idyll.
However, the results of our analysis show that less‐popular areas also attract
other groups with other motivations, such as a group of movers motivated by the
proximity to family and friends, consisting of return migrants, singles, the
youngest and oldest age groups and also the lowest‐income group. Our results
indicate that in addition to differences in rural area types, the links between
motivations and migrant characteristics are important to explaining migration
flows. Therefore, future research in popular rural areas could also use this
method of analysis to explore further the presence of different groups of migrants
with different motivations.
Comparing in‐migration flows to different types of rural areas within a
country shows that counterurbanisation does indeed entail more than what the
‘classical’ conceptualisation would suggest. Our results thus empirically support
the broader conceptualisation of counterurbanisation in the previously
mentioned model developed by Halfacree (2008), which includes groups for
whom the rural environment is important, along with groups for whom
instrumental considerations matter more. However, our results also show that
the two main groups distinguished in the model of Halfacree, mainstream and
157
default counterurbanisation, are not directly related to different types of rural
areas within countries. Future research into rural migration should be careful not
to use too simplified understandings of counterurbanisation, both in general and
with regard to different types of rural areas within countries.
6.6 More attention required for the social characteristics of rural areas
In a rural context the focus is often on the importance of landscape and locational
amenities in influencing residential choice (e.g. Argent et al., 2007). Our findings
suggest that more attention needs to be paid to the role of the perceived social
characteristics of rural areas in residential choice. In Chapter 5 it appeared that
respondents often mentioned social characteristics when referring to the
perceived characteristics they used in excluding places from the search process.
In an urban context the role of the perceived social characteristics of areas in the
residential choice process is acknowledged. The reputation of an urban
neighbourhood is known to be most strongly influenced by the socioeconomic
and ethnic composition of the neighbourhood, with physical and functional
characteristics appearing to be less important (Permentier, 2009). Furthermore, it
is assumed that relying on neighbourhood reputation in selecting
neighbourhoods during the search process can be a risk‐minimising strategy,
when information about the social quality of the area is lacking (Koopman, 2012).
In Chapter 4 we found that the social qualities of the rural environment –
such as freedom, friendliness and a pleasant atmosphere – are more important
motivations for people moving to less‐popular areas than for movers to popular
areas. This is in contrast to the popular areas, where the physical qualities of the
environment are an important pull factor. Apparently, it is useful to distinguish
between the physical and social aspects of the rural environment to explain
moving to different types of rural areas. It raises the question of whether less‐
popular areas have more to offer with regard to the social qualities of an area. To
answer this question, ‘social amenities’ will have to be looked at more closely;
however, they are less easy to operationalise than the physical characteristics
most often used to capture the amenity of rural areas (Argent et al., 2007). We
made a first attempt by using five specific dimensions of regional culture,
158
developed in a Dutch context by Brons (2005, 2006). They yielded some
differences between the three types of rural areas. However, they are very
specific to the Netherlands and the operationalisation of the dimensions is
questionable in some respects. Future research should investigate which
variables are relevant as indicators of the social characteristics of rural areas.
6.7 New methodological applications
In addition to contributing to the discussion about the reconceptualisation of
migration to rural areas, this dissertation aims to contribute to methodological
development in the field of migration and housing research. This is accomplished
by applying existing methodological approaches in a new field of research.
Firstly, whereas recent housing research has seen approaches developed which
explore the motives underlying consumer preferences (see e.g. Coolen and
Hoekstra, 2001; Jansen, 2011), underlying motivational factors such as values
have remained somewhat overlooked in migration research. Our results show
that values can add to the understanding of migration processes. They contribute
further to the characteristics and motivations of migrants when comparing
movers to different types of rural areas. Moreover, they appear to have a direct
effect on the type of area people move to, instead of interacting with or having an
effect through the motivations for choosing an area, which might also be
expected. This probably shows that these values are more closely related to the
more abstract ideas or representations that people have of areas, which are not
captured in the more concrete motivations. Apparently, the popular areas are
regarded as places to enjoy life, while the less‐popular areas represent ‘a rational
choice’ or a kind of adventurous choice. In addition, the importance of the ‘varied
life’ value for movers to less‐popular areas could indicate the existence of a kind
of group of creative people moving to these – in the Dutch context – peripheral
areas (cf. Bell and Jayne, 2010; Gibson, 2010). In this study we were only able to
measure the values with a single item, resulting in a less reliable measurement.
Based on our results, it appears worthwhile to further explore the additional
value of the use of values in migration research, using scales including more
items to measure value orientations.
159
Secondly, the diary approach we used to study the residential search
process in rural areas is a new method to study this process, but it was used
previously in other scientific fields and topics (e.g. Richardson, 1994; Laurenceau
and Bolger, 2005). Our results show that using a diary approach provides greater
insight into the residential search process, by revealing the changes in the search
space and preferences that occur between the beginning and the end of the
search. Following searchers over time shows the non‐linearity of the process and
the role played by representations. These aspects are incorporated into residential
choice models such as that of Brown and Moore (1970), but remain invisible
when using longitudinal methods with longer intervals. The qualitative elements
we used elucidate the more emotional, less tangible aspects of residential
decision‐making (see also Levy et al., 2008) and the influence of coincidence on
the process. These aspects are also overlooked when using a large quantitative
longitudinal dataset. The combination of an in‐depth interview with the
questionnaire appeared to be very useful. The personal contact and the fact that
the researcher made the effort to come and visit the respondents led to a high
level of compliance in completing the questionnaire. The interview provided rich
information, which also helped interpret the answers to the questionnaire. Due to
its more quantitative nature, the questionnaire offers the most information on the
development of the search process. The ultimate decision, involving trade‐offs
between preferences, is harder to disentangle using it. This could be solved by
adding an ‘end interview’ to the approach.
6.8 Reflection on the data and methodology
The data we used offer a rich insight into the characteristics, motivations and
search processes of rural in‐migrants to different types of rural areas. However,
our data and methodology naturally also have some relevant limitations. The
restrictions of the secondary data used in Chapter 2 and the limited measurement
of values in Chapter 4 have already been discussed in this conclusion.
In Chapter 2 we distinguish different types of rural areas using average
house price per municipality. We used the average house prices available for the
most recent year at the time, namely 2008. We used this classification of less‐
160
popular and popular areas later in the dissertation to determine the study areas
for the survey. Since 2008, the crisis in the housing market has resulted in lower
prices and decreasing numbers of transactions in most municipalities in the
Netherlands, including those in the northern Netherlands. While it is a process
that still continues, so far this has not changed the position of the less‐popular
areas with respect to the Netherlands overall, as the lowest ranked areas. The
general pattern of popularity within the North has also not changed
substantially.
We chose to use the municipal level to determine less popular and popular
rural areas and also as study areas for conducting our survey. We made this
choice because of the availability of data at the municipal level, which permitted
us to characterise the different types of rural areas in Chapter 2 and to show
migration statistics for these areas. Conducting observations at the municipal
level probably overlooks the diversity in popularity within the municipality.
However, we took this into account by choosing municipalities without large
towns within their boundaries and which were quite homogenous with regard to
their physical geography.
For the survey we selected households that had moved to the municipality
in the past five years. The chosen period needed to be as short as possible to
reduce the risk of memory‐recall problems. However, we needed to cover at least
five years to ensure sufficient completed questionnaires to conduct the analyses.
The consequence was that we failed to include people who moved to the
municipality and left again within five years. These are either very mobile people
or people who were very dissatisfied with their choice of rural area. It would be
interesting for future research to focus specifically on the characteristics and
motivations of this very mobile group.
The response rate for the less popular and popular rural areas taken
together was 24%. In the Netherlands this is considered to be a good response
rate for a postal survey. Nevertheless, if the non‐response is selective, it affects
the representativeness of the outcomes. This is not so much a problem when
investigating relationships between variables such as in Chapter 3 and 4. There is
no reason to assume that these relationships differ for people inside and outside
the sample. Furthermore, there are no indications that the pattern of non‐
161
response differs between the different types of rural areas. For the descriptive
analysis presented in Chapter 3, the effects of non‐response are more relevant.
We were able to compare the distribution of the age variable in our sample with
data from Statistics Netherlands. When compared to these data, it appeared that
in our sample the older age groups were over‐represented, while the youngest
age group was under‐represented. To obtain a representative age distribution we
weighted the sample with regard to age for the descriptive analysis.
A complement to this study would be a comparable study of less‐popular
rural areas in other countries, more specifically in more sparsely populated
countries. In comparison to other European countries, the Netherlands is densely
populated and urbanised. Our results should be generalisable to other core
regions in Europe, but as Grimsrud (2011) has shown, caution is needed when
applying conceptualisations of rural in‐migration originating in core regions in
Europe to more sparsely populated countries. However, attention is also required
in those countries for the diversity of rural in‐migration within countries.
A final reflection on our methodology considers the combination of
quantitative and qualitative methods that we used in this study. The information
gathered in the in‐depth interviews (see Chapter 5) turned out to be a valuable
addition to the quantitative secondary and survey data used in the other
chapters. The interviews provided greater insight into the process of migrating to
a rural destination and the many aspects, also less tangible ones, that play a role
in decision‐making. Future research into rural in‐migration could benefit from
combining quantitative and qualitative approaches.
6.9 Implications for policy
A number of suggestions for policymakers in less‐popular rural areas can be
formulated on the basis of the results of this research. With regard to the
characteristics of the in‐migrants, we found that a large share consisted of local
and regional migrants. This makes it worth investing in retaining these groups in
an area, instead of putting a lot of effort into attracting people from further away.
The interviews with distant house searchers also revealed how strongly social
and work‐related ties interfere with their ability to realise their moving
162
intentions. When considering the migrants’ motivations, it appeared that the
rural quality of the environment plays an important role also in migration to less
popular areas. This means that it is important to preserve qualities such as quiet,
space, nature and the landscape also in less‐popular areas, in order for them to
remain attractive for in‐migrants. In Section 6.6 we also emphasised the
importance of social qualities in residential choice. However, these social
qualities seem to be more difficult to influence with policy measures.
Chapter 5 shows that representations of areas play an important role in the
search process in different ways. Firstly, direct or indirect contact with an area
seems to be essential for including that area in the search space. Therefore,
encouraging tourism in an area can be an indirect way of attracting future
inhabitants, and also experiments such as handing out free train tickets or ‘try
living here’ events could be worthwhile for areas wishing to attract people. In the
province of Drenthe they have a yearly ‘try living here’ weekend (see
proef.drenthe.nl) during which the participants, often people living in the
Randstad area, stay one night in a hotel in the area. The programme includes
village inhabitants showing participants around and letting them experience
daily life in the village. Meeting people who have already taken the step to move
to Drenthe is also included in the itinerary.
However, knowledge of an area does not necessarily lead to a positive
appreciation of the area. Both local and regional searchers mentioned areas
within the province or specific villages that they did not want to live in, and
distant movers referred to entire provinces that were omitted from consideration
based on perceived characteristics. However, such perceived characteristics are
difficult to influence positively by policy measures. Nevertheless, it is important
not to influence them negatively. The discussion about population decline in the
Netherlands leads to images in the media of houses with boarded‐up windows
and empty shops. In some cases it seems as though policymakers at a municipal
or provincial level want to reinforce such images to strengthen their case for
receiving grants from the national government, for instance. This might not be a
good strategy for attracting new inhabitants. It is important to acknowledge that,
as in urban areas, the reputations areas have play a role in residential choice, and
that these reputations are persistent. Therefore, revitalising an area is not
163
sufficient, the image outsiders have of the area also needs to change (see also
Koopman, 2012).
Finally, Chapter 4’s results clearly show that the values people consider
important in life differ for migrants to different types of rural areas, and that
these values seem to be related to more abstract ideas or representations that
people have of the areas. Some additional research into this phenomenon would
seem worthwhile, as discussed above, and using these kinds of values in the
marketing of rural areas could offer additional opportunities, in the same way
that values are used in an urban context. The analysis revealed that less‐popular
areas represent ‘a rational choice’ or even a kind of adventurous choice: less‐
popular rural areas as locations for leading a life full of challenge, novelty and
change. In general it is important for policymakers in rural areas to acknowledge
that migration into rural areas is not just a matter of middle‐class urbanites in
search of a rural lifestyle. Based on our results it can be concluded that ‘the’ rural
migrant does not exist. It is essential to recognise that there are different groups
of migrants with different motivations and that these different groups of
migrants need to be targeted in different ways.
References
Andersen HS. 2011. Explanations for long‐distance counter‐urban migration into
fringe areas in Denmark. Population, Space and Place 17: 627‐641.
Argent N, Smailes P, Griffin T. 2007. The amenity complex: towards a framework
for analysing and predicting the emergence of a multifunctional
countryside in Australia. Geographical Research 45: 217‐232.
Bell D, Jayne, M. 2010. The creative countryside: policy and practice in the UK
rural cultural economy. Journal of Rural Studies 26: 209‐218.
Bolton N, Chalkley B. 1990. The rural population turnround: a case‐study of
North Devon. Journal of Rural Studies 6: 29‐43.
Brons LJ. 2005. Rethinking the Culture‐Economy Dialectic. Groningen University:
Groningen.
Brons LJ. 2006. Indirect measurement of regional culture in the Netherlands.
Tijdschrift voor Economische en Sociale Geografie 97: 547‐566.
164
Brown LA, Moore EG. 1970. The intra‐urban migration process: a perspective.
Geografiska Annaler 52(1): 1‐13.
Coolen H, Hoekstra J. 2001. Values as determinants of preferences for housing
attributes. Journal of Housing and the Built Environment 16: 285‐306.
De Groot, C, Daalhuizen, FBC, Van Dam, F, Mulder, CH. 2012. Once an outsider,
always an outsider? The accessibility of the Dutch rural housing market
among locals and non‐locals. Journal of Rural Studies doi:
10.1016/j.jrurstud.2011.12.001.
Fitchen JM. 1995. Spatial redistribution of poverty through migration of poor
people to depressed rural communities. Rural Sociology 60: 181‐201.
Foulkes M, Newbold KB. 2008. Poverty catchments: migration, residential
mobility, and population turnover in impoverished rural Illinois
communities. Rural Sociology 73: 440‐462.
Gibson C. 2010. Guest editorial – Creative geographies: tales from the ‘margins’.
Australian Geographer 41: 1‐10.
Gkartzios M, Scott M. 2009. Residential mobilities and house building in rural
Ireland: evidence from three case studies. Sociologia Ruralis 50: 64‐84.
Grimsrud GM. 2011. How well does the ‘counter‐urbanisation story’ travel to
other countries? The case of Norway. Population, Space and Place 17: 642‐655
Haartsen T, Venhorst V. 2010. Planning for decline: anticipating on population
decline in the Netherlands. Tijdschrift voor Economische en Sociale Geografie
101: 218‐227.
Halfacree K. 2008. To revitalise counterurbanization research? Recognising an
international and fuller picture. Population, Space and Place 14: 479‐495.
Halfacree K. 2012. Heterolocal identities? Counter‐urbanisation, second homes,
and rural consumption in the era of mobilities. Population, Space and Place
18: 209‐224.
Jansen SJT. 2011. Lifestyle method. In The Measurement and Analysis of Housing
Preference and Choice, Jansen SJT, Coolen HCCH, Goetgeluk RW (eds.).
Springer: Dordrecht; 177‐202.
Koopman, M. 2012. Economic Analysis of Neighbourhood Quality, Neighbourhood
Reputation and the Housing Market. IOS Press: Amsterdam.
165
Laurenceau J, Bolger N. 2005. Using diary methods to study marital and family
processes. Journal of Family Psychology 19(1): 86‐97.
Levy D, Murphy L, Lee CKC. 2008. Influences and emotions: Exploring family
decision‐making processes when buying a house. Housing Studies 23: 271‐
289.
Luttik J. 2000. The value of trees, water and open space as reflected by house
prices in the Netherlands. Landscape and Urban Planning 48: 161‐167.
Marlet G. 2009. De Aantrekkelijke Stad. VOC Uitgevers: Nijmegen.
McGranahan DA. 2008. Landscape influence on recent rural migration in the U.S.
Landscape and Urban Planning 85: 228‐240.
Milbourne P. 2007. Re‐populating rural studies: migrations, movements and
mobilities. Journal of Rural Studies 23: 381‐386.
Permentier, M. 2009. Reputation, Neighbourhoods and Behaviour. PhD thesis Faculty
of Geography: University of Utrecht.
Richardson A. 1994. The health diary: an examination of its use as a data
collection method. Journal of Advanced Nursing 19: 782‐791.
Stockdale A. 2006. Migration: Pre‐requisite for rural economic regeneration?
Journal of Rural Studies 22: 354‐366.
Ulrich RS. 1986. Human responses to vegetations and landscapes. Landscape and
Urban Planning 13: 29‐44.
Van Dam F, Heins S, Elbersen BS. 2002. Lay discourses of the rural and stated
and revealed preferences for rural living: some evidence of the existence of
a rural idyll in the Netherlands. Journal of Rural Studies 18: 461‐476.
Visser P, Van Dam F. 2006. De Prijs van de Plek: Woonomgeving en Woningprijs. Nai
Uitgevers/Ruimtelijk Planbureau: Rotterdam/Den Haag.
Woods M. 2005. Rural Geography. Sage Publications: London.
Woods M. 2011. Rural. Routledge: Oxon/New York.
166
167
Appendix A: Survey Vragenlijst Verhuizen [Gemeente] U ontvangt deze vragenlijst omdat u in de afgelopen vijf jaar in deze gemeente bent komen wonen. De vragen hebben betrekking op uw verhuizing naar deze gemeente en dus niet op eventuele verhuizingen daarna binnen de gemeente. In de vragenlijst wordt verschillende keren de term ‘platteland’ gebruikt. Wij bedoelen hiermee zowel de dorpen als het buitengebied in uw gemeente, dus ook de grotere plaatsen vallen voor dit onderzoek binnen het platteland. Naast gemeente gebruiken we in de vragenlijst zo nu en dan ook de term ‘gebied’. We bedoelen hiermee de gemeente en de omliggende gemeentes. De vragenlijst begint met een aantal vragen over uw huidige woonsituatie en over waar u eerder in uw leven hebt gewoond. Daarna willen we graag wat meer weten over uw beeld van het platteland en van het gebied waarin u woont in het bijzonder. Vervolgens gaan de vragen over de keuze van de woning in deze gemeente; hoe het zoeken naar deze woning in zijn werk is gegaan en welke zaken een belangrijke rol speelden in die keuze. Als laatste volgen er een aantal vragen over u en uw huishouden.
1. In wat voor type woning woont u?
Tussen- of hoekwoning Vrijstaande woning
Twee-onder-een-kapwoning Appartement Anders, nl. ___________
2. Is het een huur- of een
koopwoning?
Huurwoning Koopwoning
3. Wat is de ligging van de woning?
In de dorpskern In een woonwijk(je)
Aan de rand van het dorp Buiten de bebouwde kom
4. Wat is uw postcode? _ _ _ _ _ _
5. Wilt u in onderstaande tabel een volledig overzicht geven van alle plaatsen waar u
gewoond hebt, vanaf uw geboorte tot het moment waarop u naar deze gemeente verhuisd bent? Geef in jaren aan hoe lang u (ongeveer) in deze plaats hebt gewoond en kruis aan of het een stad of een dorp was. Indien u een partner heeft, kunt u dan hetzelfde doen voor uw partner?
Hed
en
<
----
----
---
G
eboo
rte
Uzelf Uw eventuele partner Aantal jaren
Woonplaats Stad Dorp Aantal jaren Woonplaats Stad Dorp
Huidige woonsituatie
Woongeschiedenis
168
6. Kende u voordat u naar deze gemeente verhuisde de gemeente of omliggende gemeentes al redelijk goed? (zo ja, dan kunt u meerdere antwoorden aangeven)
Nee, voordat ik naar deze gemeente verhuisde kende ik het gebied nog niet uit eigen ervaring
Ja, ik heb eerder in het gebied gewoond Ja, door bezoek aan familie of vrienden Ja, door bezoek tijdens vakantie of recreatie Ja, door werkgerelateerd bezoek Anders, nl. _________________________
We willen u nu graag een aantal vragen stellen over uw beeld van het platteland en het gebied waarin u woont in het bijzonder.
7. Welke vier woorden komen het eerst bij u op als u de term platteland hoort? Geef per
woord aan of het voor u een positieve (+), neutrale (+/-) of negatieve (-) betekenis heeft.
1 _______________ ( )
2 _______________ ( )
3 __________________ ( )
4 __________________ ( )
8. Welke vier woorden komen het eerst bij u op als u aan het gebied waarin u woont denkt?
Geef per woord aan of het voor u een positieve (+), neutrale (+/-) of negatieve (-) betekenis heeft.
1 _______________ ( )
2 _______________ ( )
3 __________________ ( )
4 __________________ ( )
9. Hebt u nu een ander beeld van het gebied dan u had voordat u naar deze gemeente
verhuisde? Zo ja, wat is er sinds uw verhuizing naar deze gemeente veranderd in uw beeld?
Nee Ja, nl.
______________________________________________________ ________________________________________________________________________________________________________________
10. Hoe lang hebt u (ongeveer) gezocht naar een andere woning voor uw verhuizing naar deze gemeente? _______________
11. Was het tijdens het zoeken naar een andere woning voor u ook een optie in de stad te
blijven of te gaan wonen? Ja Nee
12. Waar hebt u huizen bezichtigd voordat uw keuze viel op de woning in deze gemeente?
__________________________________________________________________________________________________________________________
Beelden
Zoekproces
169
13. Waarom hebt u niet voor de eerder bezichtigde woningen gekozen? Had dat met name te maken met: (u kunt meerdere antwoorden aangeven)
Kenmerken van de bezichtigde woningen
Huizenprijzen te hoog Het landschap De uitstraling van de
bebouwing in het algemeen
Te weinig voorzieningen
Afstand/bereikbaarheid Indruk van de bewoners van
het gebied Anders, nl.
_______________________________________________________________
14. Wilde u graag in dit gebied (deze gemeente en omliggende gemeentes) wonen en zocht u
daarbinnen een huis, of zocht u vooral een bepaald soort huis en ligging en had dat in verschillende gebieden kunnen staan?
Ik wilde vooral in dit gebied wonen Ik zocht vooral een bepaald soort huis en ligging Beide waren even belangrijk
15. Het zoeken naar een andere woning kan op verschillende manieren plaatsvinden. Geef
voor onderstaande manieren aan hoe belangrijk ze voor u waren bij het zoeken naar een andere woning.
Erg onbelangrijk
Onbelangrijk Neutraal Belangrijk Erg
belangrijk
Funda of andere huizensites 1 2 3 4 5
Woningbouwcorporatie 1 2 3 4 5
Rondgekeken in gebieden 1 2 3 4 5
Makelaars 1 2 3 4 5
Via familie/vrienden 1 2 3 4 5
Krant 1 2 3 4 5
Andere manier, nl. ______________________
1 2 3 4 5
16. Wat was de reden voor de verhuizing? (u kunt meerdere antwoorden aangeven)
Huwelijk of samenwonen Scheiding Zelfstandig gaan wonen Kinderen uit huis gegaan Verandering t.a.v. werk Pensioen
Vorige woning Vorige buurt Te hoge woonlasten Weg uit de stad Anders, nl.
________________
17. Waarom wilde u op het platteland (blijven) wonen? _________________________________________________________ _________________________________________________________
18. Waarom bent u juist in dit gebied gaan wonen?
__________________________________________________________ __________________________________________________________
Verhuisredenen en woonvoorkeuren
170
19. Hoe belangrijk waren onderstaande kenmerken van de woning in de keuze voor uw huis toen u in deze gemeente kwam wonen? Bij sommige kenmerken wordt een aanvullende open vraag gesteld, deze hoeft u alleen in te vullen als u het bewuste kenmerk erg belangrijk of belangrijk vindt.
Erg
onbelangrijk Onbelangrijk Neutraal Belangrijk
Erg belangrijk
Type woning (vrijstaand, tussenwoning, etc.)
1 2 3 4 5
Grootte tuin of buitenruimte 1 2 3 4 5
Grootte woning 1 2 3 4 5
Prijs van de woning 1 2 3 4 5
Ligging van de woning 1 2 3 4 5
Bouwstijl van de woning 1 2 3 4 5
Hoe zou u de bouwstijl van de woning omschrijven?
__________________________
Bouwperiode van de woning 1 2 3 4 5
Hoe oud is de woning ongeveer?
__________________________
De staat van onderhoud 1 2 3 4 5
Wat was de staat van de woning?
__________________________
Ander kenmerk, nl. ____________ 1 2 3 4 5
__________________________
171
20. Hoe belangrijk waren onderstaande kenmerken van de woonomgeving in de keuze voor uw huis toen u in deze gemeente kwam wonen? Bij sommige kenmerken wordt een aanvullende open vraag gesteld, deze hoeft u alleen in te vullen als u het bewuste kenmerk erg belangrijk of belangrijk vindt.
Erg onbelangrijk Onbelangrijk Neutraal Belangrijk
Erg belangrijk
Bereikbaarheid van werk 1 2 3 4 5
Nabijheid van grotere plaats 1 2 3 4 5
Beschikbaarheid OV 1 2 3 4 5
Aanwezigheid van winkel voor dagelijkse boodschappen in het dorp
1 2 3 4 5
Aanwezigheid van basisschool in het dorp 1 2 3 4 5
Aanwezigheid van voorzieningen in het dorp (bijv. sport, huisarts, dorpshuis)
1 2 3 4 5
Welke voorzieningen vond u met name belangrijk?
_______________________________
Soort landschap 1 2 3 4 5
Hoe zou u het landschap in uw omgeving beschrijven?
_______________________________
Nabijheid natuurgebied 1 2 3 4 5
Nabijheid water (bijv. een meer, Waddenzee) 1 2 3 4 5
Nabijheid bos 1 2 3 4 5
Het imago van het dorp of gebied 1 2 3 4 5
Veiligheid 1 2 3 4 5
Rust 1 2 3 4 5
Ruimte 1 2 3 4 5
Privacy 1 2 3 4 5
De status die het wonen in het dorp of gebied heeft 1 2 3 4 5
De mate waarin religie een rol speelt in het dorp 1 2 3 4 5
Nabijheid van familie en vrienden 1 2 3 4 5
Bewoners leken een open houding ten opzichte van ‘nieuwkomers’ te hebben
1 2 3 4 5
Er woonden veel ‘nieuwkomers’ 1 2 3 4 5
Er woonden mensen met dezelfde manier van leven als ikzelf
1 2 3 4 5
Bewoners leken onderling veel contact te hebben 1 2 3 4 5
Bewoners leken vooral hun eigen gang te gaan 1 2 3 4 5
Er leek een actief verenigingsleven te zijn 1 2 3 4 5
Ander kenmerk, nl. _______________________________ 1 2 3 4 5
172
21. De vorige vraag gaat over de kenmerken die u belangrijk vond toen u de keuze maakte voor de woning in deze gemeente. Zijn sinds u in de gemeente woont bepaalde aspecten van het wonen in de gemeente u meegevallen en zo ja, welke? _______________________________________________________________________________________________________________________________________________________________________________________
22. Zijn sinds u in de gemeente woont bepaalde aspecten van het wonen in de gemeente u
tegengevallen en zo ja, welke? _______________________________________________________________________________________________________________________________________________________________________________________
23. Geef door middel van een rapportcijfer aan hoe tevreden u bent u met het wonen in deze
gemeente (1-10). _____
24. Verwacht u te gaan verhuizen binnen de komende twee jaar?
Vrijwel zeker niet ----------> Ga door naar vraag 27 Misschien Ja
25. Waarom verwacht u binnen twee jaar te gaan verhuizen?
_____________________________________________________________ _____________________________________________________________
26. Waar verwacht u dan naartoe te verhuizen?
Binnen mijn huidige woonplaats
Binnen de gemeente
Binnen Noord-Nederland Naar elders in Nederland Weet nog niet
Als afsluiting van de vragenlijst willen we u nog graag een aantal vragen stellen over u en uw huishouden.
27. Wat is uw geslacht?
Man Vrouw
28. In welk jaar bent u geboren? ____________
29. Wat is uw hoogst voltooide opleiding? lagere school, basisschool LBO (LTS, LEAO, LHNO,
etc.) MAVO, MULO, VMBO HAVO, MBO
VWO, Atheneum, Gymnasium, HBS
HBO/Universitaire opleiding Anders, nl. _____________
30. Wat is uw belangrijkste vervoermiddel? Auto Openbaar vervoer Fiets
Brommer/scooter Motor Anders, nl. _____________
31. Over hoeveel auto’s beschikt uw huishouden? ________
32. Hoe is uw huishouden samengesteld?
Achtergrondkenmerken
173
Eenpersoonshuishouden Ga naar vraag 34 Echtpaar/vaste partners zonder kinderen Eén ouder met thuiswonende kinderen Echtpaar/vaste partners met thuiswonende kinderen Andere samenstelling
33. Wat is/zijn de leeftijd(en) van uw thuiswonende kinderen?
___,___,___,___,___,___,___
34. Kunt u aangeven in welke klasse het netto maandinkomen van uw huishouden (uzelf + evt. partner of andere inwonende gezinsleden valt)?
minder dan € 1250 per maand € 1250 tot € 2500 per maand € 2500 tot € 4000 per maand meer dan € 4000 per maand
35. Wat is uw belangrijkste bron van inkomsten?
Loondienst Inkomsten uit eigen bedrijf Uitkering AOW/(Pre)pensioen Ga naar vraag 40 Anders,nl. _______________________
36. Wat voor werk doet u?
____________________________________
37. In welke sector bent u werkzaam? Landbouw en visserij Delfstoffenwinning Industrie Nutsbedrijven Bouw Handel en reparatie Horeca Vervoer en communicatie Financiële instellingen Zakelijke dienstverlening Openbaar bestuur Onderwijs Zorg Overige dienstverlening Anders, nl.
_____________
174
38. In welke plaats werkt u? ____________________ 39. Werkt u overdag regelmatig thuis?
Nee, nooit Minder dan 1 dag per week Minimaal 1 dag per week Meestal
40. Indien u een partner heeft: werkt uw partner? Nee Ja, in deze plaats: ___________________
41. Naast deze achtergrondkenmerken hebben mensen verschillende persoonlijke en sociale
waarden die ze belangrijk vinden in hun leven. Hieronder staan een aantal van deze waarden genoemd. Geef voor elk van deze waarden aan hoe belangrijk deze is in uw leven.
Erg onbelangrijk Onbelangrijk
Gemiddeld belangrijk Belangrijk Erg belangrijk
Vrijheid (in denken en doen) 1 2 3 4 5
Plezier in uw leven (genot, voldoening)
1 2 3 4 5
Een harmonieus gezinsleven
1 2 3 4 5
Zelfdiscipline (zelfbeperking, bestand tegen verleidingen)
1 2 3 4 5
Rijkdom (materiële bezittingen, geld)
1 2 3 4 5
Rationeel zijn (intelligent, logisch, nadenkend)
1 2 3 4 5
Uw imago bij anderen in stand houden (gezichtsverlies voorkomen)
1 2 3 4 5
Afwisseling (een leven vol met uitdaging, nieuwigheid en verandering)
1 2 3 4 5
42. Bent u bereid mee te werken aan een vervolginterview? Zo ja, wilt u uw telefoonnummer
of emailadres noteren zodat wij eventueel contact met u kunnen opnemen hiervoor? ________________________________________________________________________________________________________________________
43. Heeft u nog vragen of opmerkingen over de vragenlijst? ____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
Dit is het einde van de vragenlijst. Heel erg bedankt voor uw tijd en moeite.
Overig
175
Appendix B: In‐depth interview guide
Startinterview dagboekbenadering
Algemeen
- Hoe is uw huishouden samengesteld?
- Hoe lang woont u al in deze plaats?
- Kunt u vertellen waar u en uw partner vanaf de geboorte tot nu hebben gewoond?
- Heeft u, naast misschien door wonen, nog op een andere manier een bepaalde affiniteit of
contact met bepaalde landelijke gebieden in Noord‐Nederland? Met welke gebieden is
dat?
- Werken u en uw partner?
- Zo ja, wat voor werk doet u en wat voor werk doet uw partner? (als met pensioen: welk
werk hiervoor gedaan?)
- Verwacht u voor u of uw partner na de verhuizing een verandering wat betreft werk?
- Bent u door uw werk of andere factoren gebonden om in de buurt van een bepaalde
plaats of in een bepaald gebied in Noord‐Nederland te wonen?
Zoeken naar andere woning
- Waarom wilt u verhuizen?
- Hoe lang geleden bent u begonnen met zoeken naar een ander huis?
- Toen u begon met zoeken naar een andere woning, in welk gebied zocht u toen? Waarom
in dit gebied?
- Op welke manier heeft u uw zoekgebied vervolgens aangepast? Wat was daarvoor de
aanleiding?
- Wat is op dit moment het gebied waarin u naar huizen zoekt?
- Is in plaats van platteland wonen in de stad ook een mogelijkheid?
- Wilt u graag in dit gebied wonen en zoekt u daarbinnen een huis, of zoekt u vooral een
bepaald soort huis en plek, en zou dat in verschillende gebieden kunnen staan?
- Op welke manieren hebt u tot nu toe gezocht?
- Waarom wilt u graag op het platteland (blijven) wonen?
176
- Wat maakt dan in het bijzonder het gebied wat u zojuist noemde (zoekgebied)
aantrekkelijk voor u om te wonen?
- Wat ziet u als nadelen van dit gebied?
- Denken u en uw partner overwegend hetzelfde over waar u zou willen wonen en wat
belangrijk is in die keuze voor een plek?
- Wat zijn wat dit betreft de verschillen tussen u beide?
- Is er een bepaalde termijn waarbinnen u verhuisd wilt zijn?
Beeld platteland en plattelandsgebieden in Noord‐Nederland
- Welke vier woorden komen er het eerst bij u op als u de term platteland hoort?
- U hebt zo straks verteld in welk gebied u naar huizen zoekt. Als we naar de kaart kijken,
in welke gebieden bevindt uw zoekgebied zich dan?
- Welke vier woorden komen er het eerst bij u op als u [één van de gebieden die binnen het
zoekgebied vallen] hoort? ‐> deze vraag voor ieder genoemd gebied stellen
- Welke gebieden zouden (bijv. qua ligging) ook in aanmerking kunnen komen?
- Wat is uw beeld van deze gebieden?
Woonvoorkeuren
‐ Ik wil graag weten welke kenmerken u van belang vindt voor uw volgende
woonomgeving en woning. Ik onderscheid daarbij 5 thema’s: het landschap,
voorzieningen, sociale kenmerken, locatie en kenmerken van de woning. Het is de
bedoeling dat u per thema kenmerken noemt die u belangrijk vind in de keuze voor een
nieuw huis.
‐ Kenmerken van het landschap (bijv. soort landschap, nabijheid natuurgebied, nabijheid
water, nabijheid bos)
‐ Voorzieningen (bijv. supermarkt, basisschool, kinderopvang, huisarts, dorpshuis,
sportvoorzieningen, culturele voorzieningen, openbaar vervoer)
‐ Sociale kenmerken (bijv. privacy, religie, actief verenigingsleven,
bevolkingssamenstelling, het imago van een dorp of gebied, de status die het wonen in
een dorp of gebied heeft, houding bewoners t.o.v. nieuwkomers, mate van onderling
contact van bewoners)
‐ Locatie (bijv. grootte van de plaats, buitengebied of dorp, bereikbaarheid werk, nabijheid
grotere plaats)
177
‐ Kenmerken woning (bijv. type woning, aanwezigheid bijgebouwen, vormgeving woning,
bouwperiode, aantal kamers, oppervlakte woning, oppervlakte perceel, staat van
onderhoud, ligging)
‐ Tenslotte zou ik nog graag willen dat u een rangorde aanbrengt in deze verschillende
thema’s, hoe belangrijk zijn ze voor u in de keuze voor een nieuwe woning?
Aanvullend schriftelijk vragenlijstje (ter plekke laten invullen)
In welk jaar bent u geboren?
…………………….
In welk jaar is uw partner geboren?
……………………
Wat is de leeftijd van uw kinderen?
….., ….., ….., ……, ….., ….., ….., …..,
Wat is uw hoogst voltooide opleiding?
1 lagere school, basisschool
2 LBO (LTS, LEAO, LHNO, etc.)
3 MAVO, MULO
4 VMBO
5 HAVO
6 MBO
7 VWO, Atheneum, Gymnasium, HBS
8 HBO
9 Universitaire opleiding
10 Anders, nl. …
Kunt u aangeven in welke klasse het netto maandinkomen van uw huishouden (uzelf + partner
valt)?
1 minder dan € 1250 per maand
2 € 1250 tot € 2000 per maand
3 € 2000 tot € 3000 per maand
4 € 3000 tot € 4000 per maand
5 meer dan € 4000 per maand
Over hoeveel auto’s heeft uw huishouden de beschikking?
…
Kunt u aangeven wat ongeveer de maximale prijs is die u voor uw nieuwe huis wilt betalen?
…
179
Appendix C: Electronic questionnaire
Elektronische vragenlijst dagboekbenadering
De respondenten ontvingen iedere twee weken via email een link naar deze elektronische
vragenlijst. Doordat ze inlogden met een persoonlijke code konden ze bij sommige vragen zien
wat ze de keer ervoor hadden ingevuld. Daarnaast biedt een elektronische vragenlijst de
mogelijkheid op basis van het antwoord op de voorgaande vraag bepaalde vragen wel of niet te
tonen op het scherm. In deze schriftelijke weergave is dat zo goed mogelijk geprobeerd weer te
geven.
Verhuizen Noord‐Nederland
Welkom bij de tweewekelijkse vragenlijst over uw zoektocht naar een nieuwe woning. De vragen gaan over
wat u in de afgelopen twee weken hebt ondernomen met betrekking tot het zoeken van een nieuwe woning
en uw ideeën over uw toekomstige woning en woonomgeving.
Alvast erg bedankt voor het invullen van de vragen!
Heeft u in de afgelopen twee weken een huis gekocht?
○ Ja ○ Nee
Indien huis gekocht dan verschijnen deze vragen:
In welke plaats staat dit huis?
‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐
Welke kenmerken van het huis waren belangrijk in uw beslissing om dit huis te kopen?
U kunt meerdere kenmerken kiezen. Geef voor de gekozen kenmerken in de open ruimte aan wat
dit kenmerk in dit geval inhoudt. Als voor u bijvoorbeeld ‘ligging’ een belangrijk kenmerk is,
kunt u hier aangeven wat de ligging van het huis is.
○ Type woning ______________________________
○ Grootte perceel ______________________________
○ Aantal kamers ______________________________
○ Aanwezigheid bijgebouwen ______________________________
○ Prijs van de woning ______________________________
○ Vormgeving van de woning ______________________________
○ Bouwperiode van de woning ______________________________
○ Staat van onderhoud ______________________________
○ Ligging ______________________________
○ ander kenmerk, nl.: ______________________________
180
Welke kenmerken van de woonomgeving waren belangrijk in uw beslissing om dit huis te
kopen?
U kunt meerdere kenmerken kiezen. Sommige kenmerken spreken meer voor zich dan andere. In
de open ruimte kunt u de door u gekozen kenmerken eventueel toelichten. Als u bijvoorbeeld
‘soort landschap’ een belangrijk kenmerk vindt, kunt u hier aangeven welk soort landschap zich
in de woonomgeving bevindt.
○ Bereikbaarheid van werk
○ Nabijheid van grotere plaats
○ Aanwezigheid van winkel voor dagelijkse boodschappen in het dorp
○ Aanwezigheid van basisschool in het dorp
○ Aanwezigheid van andere voorzieningen (bijv. kinderopvang, dorpshuis, huisarts,
sportvoorzieningen)
○ Aanwezigheid OV
○ Het is een relatief groot dorp
○ Het is een relatief klein dorp
○ Soort landschap
○ Nabijheid water
○ Nabijheid bos
○ Aantrekkelijke bebouwing
○ Het imago van een dorp of gebied
○ Veiligheid
○ Rust ○ Privacy
○ De status die het wonen in het dorp of gebied heeft
○ Het dorp is niet‐kerkelijk
○ Religie speelt een belangrijke rol in het dorp ○ Nabijheid van familie en vrienden
○ Bewoners hebben een open houding ten opzichte van ‘nieuwkomers’
○ Er wonen veel ‘nieuwkomers’
○ Er wonen mensen met dezelfde manier van leven als ikzelf
○ Bewoners hebben onderling veel contact
○ Bewoners hebben onderling weinig contact
○ Actief verenigingsleven
○ ander kenmerk, nl.:
Dan: Einde vragenlijst
Indien geen huis gekocht dan verschijnt deze vraag:
Bent u nog op zoek naar een ander huis?
○ Ja ○ Nee, definitief gestopt met zoeken
181
Indien ‘Nee, definitief gestopt met zoeken’ dan verschijnt deze vraag:
Waarom bent u gestopt met zoeken?
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
Dan: Einde vragenlijst
Indien ja (nog steeds op zoek) dan verschijnt deze vraag:
Heeft u in de afgelopen twee weken huizen bezichtigd?
○ Ja
○ Nee
Alleen als het antwoord ja is verschijnen deze vragen over de bezichtigingen:
In welke plaats(en) bevindt dit huis / bevinden deze huizen zich?
Huis 1: Bijv. Groningen_________________________________
Huis 2: ______________________________________________
Huis 3: ______________________________________________
Huis 4: ______________________________________________
Per huis verschijnen de volgende twee vragen:
Wat ziet u na het bezichtigen van huis 1 in [Groningen] als positieve aspecten van de woning en
de woonomgeving?
______________________________________________________________________________________
______________________________________________________________________________________
__________________________________________________________________________________
Wat ziet u na het bezichtigen van huis 1 in [Groningen] als negatieve aspecten van de woning en
de woonomgeving?
______________________________________________________________________________________
______________________________________________________________________________________
___________________________________________________________________________________
182
De volgende vragen verschijnen voor iedereen:
Als u de vragenlijst voor de eerste keer invult geef dan aan hoe belangrijk u onderstaande
kenmerken vindt voor uw toekomstige woning.
Als u de vragenlijst eerder hebt ingevuld hebt u aangegeven hoe belangrijk u onderstaande
kenmerken vindt voor uw toekomstige woning. Pas uw antwoord(en) aan als het belang van een
of meerdere kenmerken voor u veranderd is. Voor de kenmerken waarvan het belang voor u
gelijk is gebleven kunt u de antwoorden ongewijzigd laten.
Erg belangrijk Belangrijk Neutraal Onbelangrijk Zeer onbelangrijk Geen antwoord
Type woning ○ ○ ○ ○ ○ ○
Grootte perceel ○ ○ ○ ○ ○ ○
Aantal kamers ○ ○ ○ ○ ○ ○
Aanwezigheid ○ ○ ○ ○ ○ ○
bijgebouwen
Prijs van de woning ○ ○ ○ ○ ○ ○
Vormgeving van de ○ ○ ○ ○ ○ ○
woning
Bouwperiode van de ○ ○ ○ ○ ○ ○
woning
Staat van onderhoud ○ ○ ○ ○ ○ ○
Ligging ○ ○ ○ ○ ○ ○
Op de volgende pagina worden de kenmerken weergegeven die u belangrijk vindt voor uw
toekomstige woning. Geef voor de kenmerken waarbij nog niets staat ingevuld aan welke
invulling deze volgens u zouden moeten hebben.
Bekijk voor de kenmerken waarbij al iets staat ingevuld of de invulling nog steeds overeenkomt
met uw huidige ideeën. Als dit niet het geval is, pas de invulling van het kenmerk dan aan
(bijvoorbeeld: uw voorkeur gaat ten aanzien van de bouwperiode in plaats van naar vooroorlogs
nu uit naar nieuwbouw).
Type woning _________________________
Grootte perceel _________________________
Aantal kamers _________________________
Aanwezigheid bijgebouwen _________________________
Prijs van de woning _________________________
Vormgeving van de woning _________________________
Bouwperiode van de woning _________________________
Staat van onderhoud _________________________
Ligging ________________________
183
Als er iets veranderd is in de kenmerken die u belangrijk vindt of in de gewenste invulling
daarvan, kunt u dan aangeven wat de aanleiding was voor deze verandering(en)?
______________________________________________________________________________________
______________________________________________________________________________________
Als u de vragenlijst voor de eerste keer invult geef dan aan hoe belangrijk u onderstaande
kenmerken vindt voor uw toekomstige woonomgeving.
Als u de vragenlijst eerder hebt ingevuld hebt u aangegeven hoe belangrijk u onderstaande
kenmerken vindt voor uw toekomstige woonomgeving. Pas uw antwoord(en) aan als het belang
van een of meerdere kenmerken voor u veranderd is. Voor de kenmerken waarvan het belang
voor u gelijk is gebleven kunt u de antwoorden ongewijzigd laten.
(net als bij de vraag op de vorige bladzijde kon hierbij gekozen worden uit de opties: Erg belangrijk,
Belangrijk, Neutraal, Onbelangrijk, Zeer onbelangrijk, Geen antwoord)
Bereikbaarheid van werk
Nabijheid van grotere plaats
Aanwezigheid van winkel voor dagelijkse boodschappen in het dorp
Aanwezigheid van basisschool in het dorp
Aanwezigheid van andere voorzieningen (bijv. kinderopvang, dorpshuis, huisarts, sportvoorzieningen)
Aanwezigheid OV
Het is een relatief groot dorp
Het is een relatief klein dorp
Soort landschap
Nabijheid water
Nabijheid bos
Aantrekkelijke bebouwing
Het imago van een dorp of gebied
Veiligheid
Rust
Privacy
De status die het wonen in het dorp of gebied heeft
Het dorp is niet‐kerkelijk
Religie speelt een belangrijke rol in het dorp
Nabijheid van familie en vrienden
Bewoners hebben een open houding ten opzichte van ‘nieuwkomers’
Er wonen veel ‘nieuwkomers’
Er wonen mensen met dezelfde manier van leven als ikzelf
Bewoners hebben onderling veel contact
Bewoners hebben onderling weinig contact
Actief verenigingsleven
184
Op de volgende pagina worden de kenmerken weergegeven die u belangrijk vindt voor uw
toekomstige woonomgeving. Voor de kenmerken waarbij nog niets staat ingevuld kunt u
eventueel toelichten welke invulling deze volgens u zouden moeten hebben. Bekijk voor de
kenmerken waarbij al iets staat ingevuld of de invulling nog steeds overeenkomt met uw huidige
ideeën. Als dit niet het geval is, pas de invulling van het kenmerk dan aan (bijvoorbeeld: uw
voorkeur gaat nu uit naar een ander soort landschap dan eerder).
Bereikbaarheid van werk
Nabijheid van grotere plaats
Aanwezigheid van winkel voor dagelijkse boodschappen in het dorp
Aanwezigheid van basisschool in het dorp
Aanwezigheid van andere voorzieningen (bijv. kinderopvang, dorpshuis, huisarts, sportvoorzieningen)
Aanwezigheid OV
Het is een relatief groot dorp
Het is een relatief klein dorp
Soort landschap
Nabijheid water
Nabijheid bos
Aantrekkelijke bebouwing
Het imago van een dorp of gebied
Veiligheid
Rust
Privacy
De status die het wonen in het dorp of gebied heeft
Het dorp is niet‐kerkelijk
Religie speelt een belangrijke rol in het dorp
Nabijheid van familie en vrienden
Bewoners hebben een open houding ten opzichte van ‘nieuwkomers’
Er wonen veel ‘nieuwkomers’
Er wonen mensen met dezelfde manier van leven als ikzelf
Bewoners hebben onderling veel contact
Bewoners hebben onderling weinig contact
Actief verenigingsleven
Als er iets veranderd is in de kenmerken die u belangrijk vindt of in de gewenste invulling
daarvan, kunt u dan aangeven wat de aanleiding was voor deze verandering(en)?
______________________________________________________________________________________
______________________________________________________________________________________
___________________________________________________________________________________
185
Als u de vragenlijst voor het eerst invult geef dan hieronder aan in welk gebied u naar huizen
zoekt.
Als u de vragenlijst eerder hebt ingevuld staat hieronder hoe u het gebied waarin u naar huizen
zoekt omschreven hebt.
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
Heeft u uw zoekgebied aangepast in de afgelopen twee weken?
○ Ja ○ Nee
Indien zoekgebied aangepast dan verschijnen de volgende vragen:
Wat is het totale gebied waarin u nu naar huizen zoekt? [het antwoord op deze vraag verschijnt
dan als de vragenlijst weer wordt ingevuld bij de ‘zoekgebiedvraag’ hierboven]
______________________________________________________________________________________
______________________________________________________________________________________
Wat was de aanleiding voor de aanpassing van uw zoekgebied?
______________________________________________________________________________________
______________________________________________________________________________________
In welke van de onderstaande gebied(en) bevindt uw nieuwe zoekgebied zich? (als u op de link
klikt ziet u een kaartje waarop de gebieden worden weergegeven)
○ Noordoost Friesland
○ Noordwest Friesland
○ West‐ en midden‐ friesland
○ Zuidwest Friesland
○ Zuidoost Friesland
○ Friese Wouden
○ Westerkwartier
○ Noord Groningen
○ Slochteren en omstreken
○ Oldambt
○ Westerwolde
○ Veenkoloniën
○ Noord Drenthe
○ Centraal Drenthe
○ Zuidwest Drenthe
186
Per aangevinkt gebied verschijnt de volgende vraag:
Hoe zou u het gebied [bijv. Noordoost Friesland] omschrijven?
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
Deze vragen verschijnen voor iedereen:
Heeft u in de afgelopen twee weken een of meerdere van onderstaande zoekactiviteiten
ondernomen?
○ Geen zoekactiviteiten ondernomen
○ Naar huizen gezocht op Funda of een vergelijkbare huizensite
○ Contact met een makelaar gehad
○ Rondgekeken in landelijke gebieden waar u mogelijk zou willen wonen
○ Andere zoekactie: __________________________________________
Indien contact met een makelaar:
Waaruit bestond dit contact met een makelaar?
○ Verzoek om informatie
○ Ingeschreven als zoeker
○ Verzoek om mee te gaan bij bezichtiging
○ Aanvraag van bezichtiging bij verkopende makelaar
○ Anders, nl.: _________________
In welke plaats bevindt deze makelaar zich?
___________________________________
187
Indien rondgekeken:
In welke gebied(en) hebt u rondgekeken?
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
Had u tijdens dit zoekproces al eerder in deze gebieden rondgekeken?
○ Nee ○ Ja
Niet eerder rondgekeken:
Kunt u per gebied een korte beschrijving van uw indruk van het gebied geven?
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
Wel eerder rondgekeken:
Heeft u nieuwe indrukken opgedaan in deze gebieden?
○ Ja ○ Nee
Indien ja:
Kunt u deze nieuwe indrukken (evt. per gebied) kort beschrijven?
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
188
Voor iedereen:
Zijn er zaken met betrekking tot uw zoektocht naar een nieuw huis die niet in de vragenlijst naar
voren zijn gekomen die u nog graag wilt noemen?
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
Heeft u nog vragen of opmerkingen over de vragenlijst?
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
Dit is het einde van de vragenlijst.
Heel erg bedankt voor uw medewerking!
189
Summary in Dutch
Migratie naar minder populaire plattelandsgebieden
De kenmerken, motieven en het zoekproces van verhuizers
Achtergrond, doel en methode
Het dominante beeld van migratie naar het platteland is dat van de welvarende
stedeling op zoek naar een nieuw leven in een idyllische plattelandsomgeving.
De wetenschappelijke belangstelling voor migratie naar het platteland, waarvoor
vaak de term counterurbanisatie gebruikt wordt, begon in de jaren ’70 van de
vorige eeuw en resulteerde in een grote hoeveelheid onderzoeksresultaten.
Recentelijk is er in de wetenschappelijke literatuur een hernieuwde aandacht
waarbij het verschijnsel kritisch tegen het licht wordt gehouden. Deze dissertatie
levert een bijdrage aan deze kritische herziening van het verschijnsel migratie
naar het platteland.
Eén aspect van dit opnieuw bekijken van verhuizen naar het platteland is
de toenemende erkenning dat plattelandsgebieden verschillen in hun
populariteit als woongebied, mede onder invloed van de aantrekkelijkheid van
het landschap en bereikbaarheid. Ook is vastgesteld dat naast de
aantrekkingskracht van de rurale idylle ook andere motieven een rol spelen,
zoals de nabijheid van familie en vrienden of aan werk gerelateerde redenen.
Naast de eerder genoemde middenklasse blijken ook groepen mensen met lage
inkomens en arbeidsmigranten naar het platteland te trekken en het blijkt dat
naast migratie vanuit de stad ook migratie binnen het platteland van belang is.
Hoewel dus steeds vaker wordt erkend dat plattelandsgebieden verschillen
in hun populariteit als woongebied is dit idee nog niet terug te vinden in het
onderzoek naar de kenmerken en motieven van verhuizers naar het platteland.
Dergelijk onderzoek richt zich vaak op het platteland in het algemeen of op die
plattelandsgebieden die erg aantrekkelijk zijn om te gaan wonen. Migratie naar
minder populaire plattelandsgebieden is daarbij onderbelicht gebleven. Door
migratie naar minder populaire plattelandsgebieden in Noord‐Nederland te
onderzoeken wil deze studie een bijdrage leveren aan de ontwikkeling van een
190
bredere en genuanceerdere opvatting van counterurbanisatie met een empirische
basis. Dit is niet alleen wetenschappelijk relevant, maar het levert ook
interessante informatie op voor beleidsmakers. Net als andere
plattelandsgebieden in Europa heeft het Nederlandse platteland, vooral in de
meer perifere gebieden waaronder Noord‐Nederland, te maken met
bevolkingskrimp. In‐migratie is één van de factoren die van invloed zijn op de
bevolkingsontwikkeling, naast geboorte, sterfte en uit‐migratie. Daarom is het
belangrijk meer te weten over de kenmerken, motieven en zoekstrategieën van
mensen die naar dit soort gebieden verhuizen en met name over de verschillende
migratiestromen naar verschillende soorten plattelandsgebieden.
Het doel van deze dissertatie is om meer inzicht te krijgen in migratie naar
minder populaire plattelandsgebieden. Om te bepalen in hoeverre de gevonden
uitkomsten specifiek zijn voor minder populaire gebieden is een directe
vergelijking met verhuizers naar populaire gebieden in dezelfde regio gemaakt.
De populariteit van plattelandsgebieden als woongebied wordt in dit onderzoek
vastgesteld aan de hand van de gemiddelde huizenprijs per gemeente. Binnen
Nederland is het Noorden gekozen als studiegebied, omdat het als het meest
landelijke deel van Nederland kan worden gezien aan de hand van de
omgevingsadressendichtheid en de perceptie van de Nederlandse bevolking. Op
basis van het doel van het onderzoek zijn drie onderzoeksvragen geformuleerd:
1. Wat zijn de kenmerken en motieven van verhuizers naar minder populaire
plattelandsgebieden?
2. In hoeverre verschillen de kenmerken en motieven van verhuizers naar minder
populaire en populaire plattelandsgebieden?
3. Hoe zoeken potentiële verhuizers naar een plek om te wonen in
plattelandsgebieden?
Om deze onderzoeksvragen te beantwoorden is gebruik gemaakt van een
combinatie van verschillende soorten data en methodes. De kenmerken en
motieven van verhuizers zijn in eerste instantie onderzocht door gebruik te
maken van secundaire data afkomstig van het CBS en WoON. Uit deze
191
verkenning bleek dat aanvullende data nodig waren om de onderzoeksvragen op
een goede manier te beantwoorden, zo ontbrak informatie over terugmigratie en
waren motieven voor vertrek vermengd met motieven voor de keuze van de
nieuwe woonlocatie. Daarom hebben we zelf data verzameld door middel van
een schriftelijke enquête onder nieuwe inwoners van vier gemeenten die in
hoofdstuk 2 als minder populair geclassificeerd waren en drie populaire
gemeenten (N = 1717). Deze dataverzameling kon plaatsvinden door de goede
samenwerking met de betreffende gemeenten.
Om het zoekproces naar een nieuwe woning op het platteland te
onderzoeken is een kleinschalige studie uitgevoerd die gebruikmaakt van een
zogenaamde dagboekbenadering, een nieuwe methode om het zoekproces naar
een nieuwe woning te onderzoeken. Deze dagboekbenadering bestond uit
diepte‐interviews die gevolgd werden door een elektronische vragenlijst die
iedere twee weken per mail naar de respondenten gestuurd werd, totdat zij een
huis kochten of stopten met zoeken.
Minder populaire en populaire plattelandsgebieden in Noord‐Nederland
De populariteit van plattelandsgebieden als woongebied wordt in dit onderzoek
gedefinieerd aan de hand van de gemiddelde huizenprijs per gemeente.
Huizenprijzen geven een indicatie van wat mensen bereid zijn om te betalen voor
het huis en zijn omgeving. De gemiddelde huizenprijs omvat zowel
waarderingsaspecten van de woonomgeving als van de woning zoals grootte en
kwaliteit. Waar in andere (buitenlandse) onderzoeken vaak in‐migratiecijfers
worden gebruikt om de populariteit van plattelandsgebieden te bepalen, is in de
Nederlandse context het gebruik van huizenprijzen meer geschikt. In Nederland
wordt bevolkingsgroei op gemeenteniveau sterk bepaald door overheidsbeleid.
Nieuwbouw wordt slechts in een beperkt aantal gebieden toegestaan, terwijl in
andere gebieden het bouwen van nieuwe woningen sterk beperkt wordt. Dit
laatste geldt voor veel plattelandsgebieden. Deze beperkingen zorgen ervoor dat
bij een stijgende vraag naar huizen de prijzen toenemen in plaats van de
aantallen in‐migranten. Huurprijzen zijn niet gebruikt bij het bepalen van de
populariteit van plattelandsgebieden in dit onderzoek, omdat deze weinig
verschillen tussen gebieden in Nederland.
192
Op basis van huizenprijzen hebben we in hoofdstuk 2 drie typen
plattelandsgebieden onderscheiden in Noord‐Nederland: minder populaire
gebieden, gemiddelde gebieden en populaire gebieden. Deze gebieden worden
als platteland beschouwd op basis van de maat die het CBS gebruikt voor de
stedelijkheid van een gebied, de omgevingsadressendichtheid. Ze bestaan uit
gemeenten met gemiddeld minder dan 1000 adressen per vierkante kilometer. In
deze dissertatie staan de minder populaire gebieden centraal, maar er wordt ook
een vergelijking met de populaire gebieden gemaakt. De minder populaire
gebieden liggen langs de noordelijke en oostelijke grens van Noord‐Nederland.
Ze worden gekenmerkt door een open zeekleilandschap met een groot aandeel
grootschalige akkerbouw. Dit komt overeen met bevindingen uit eerder
onderzoek dat mensen in het algemeen juist een voorkeur hebben voor gebieden
met een afwisseling van bos en open stukken en een klein aandeel bouwland. In
vergelijking met de populaire gebieden hebben de minder populaire gebieden
minder natuur‐ en recreatiegebieden. De minder populaire gebieden hebben een
vergelijkbare reisafstand naar de belangrijke steden in het noorden, maar liggen
verder weg van het midden van Nederland dan de populaire gebieden. Voor
deze minder populaire gebieden worden zowel bevolkings‐ als
huishoudenskrimp voorspeld, in sommige gemeenten is de bevolking al aan het
afnemen. Hier tegenover staan de populaire gebieden die juist gekenmerkt
worden door een semi‐open landschap met overwegend veeteelt. Ze hebben ook
een relatief groter aandeel natuurgebieden. Verder is voor deze gebieden
huishoudensgroei voorspeld en liggen ze minder ver van het midden van
Nederland.
Als we aan de hand van migratiestatistieken van het CBS naar Noord‐
Nederland als geheel kijken blijkt dat het migratiesaldo negatief is. In de periode
2003‐2007 zijn gemiddeld 22.326 mensen uit het noorden vertrokken naar elders
in Nederland, terwijl 21.913 mensen zich er vestigden. Van deze in‐migranten
vanuit elders in Nederland ging 54% in één van de 58 gemeenten wonen die als
platteland geclassificeerd kunnen worden. Als we in dezelfde periode naar
migratie binnen de regio kijken blijkt dat 31.374 mensen (61% van de totale groep
verhuizers tussen gemeenten in Noord‐Nederland) naar een plattelandsgemeente
193
verhuisd zijn, afkomstig uit zowel stedelijke als plattelandsgemeenten. De drie
typen plattelandsgebieden, de minder populaire, gemiddelde en populaire
gebieden, laten allemaal een negatief migratiesaldo zien voor de periode 2003‐
2007 en de migratiesaldi verschillen niet significant tussen de drie typen
gebieden. Er zijn dus meer mensen vertrokken uit de gebieden dan ernaartoe
verhuisd zijn, ook in populaire gebieden. Het gemiddelde aantal in‐migranten
per 1.000 inwoners is eveneens vergelijkbaar voor de drie gebieden. Populaire
gebieden trekken dus niet significant meer mensen aan dan minder populaire
gebieden. Er vindt een aanzienlijke migratie plaats tussen stedelijke en populaire
gebieden, terwijl migratie tussen minder populaire en gemiddelde gebieden aan
de ene kant en populaire gebieden aan de andere kant erg beperkt is.
Wie verhuizen er naar minder populaire gebieden en waarom?
De uitkomsten die in hoofdstuk 3 gepresenteerd worden op basis van nieuw
verzamelde enquêtegegevens laten zien dat de verhuizers naar minder populaire
plattelandsgebieden in Noord‐Nederland meer divers zijn dan de middenklasse
die vaak geassocieerd wordt met verhuizen naar het platteland. Ze zijn eveneens
meer divers dan de verhuizers naar verarmde of krimpende plattelandsgebieden
die in buitenlandse studies gevonden zijn. Hoewel de verhuizers in de
Nederlandse context een relatief laag inkomen hebben is het aandeel van de
laagste inkomensgroep slechts klein. Gecombineerd met het relatief grote aandeel
werkenden, de variatie aan sectoren waarin ze werkzaam zijn en het grote
aandeel hoger opgeleiden lijken deze verhuizers meer op de diverse groep
verhuizers zoals die ook gevonden werd in perifere gebieden in Denemarken.
Naast deze sociaaleconomische kenmerken laten onze resultaten zien dat de
verhuizers naar minder populaire gebieden vaak jong zijn, terwijl een relatief
klein deel vanuit de stad verhuisd is. Het soort verhuizing varieert van lokaal
(vanuit omliggende gemeentes) tot over grotere afstand (vanuit elders in
Nederland), maar de grootste groep komt vanuit elders in Noord‐Nederland.
Het klassieke beeld is dat de kwaliteit van de woonomgeving het
belangrijkste motief is om naar het platteland te verhuizen. Uit andere
onderzoeken is gebleken dat voor verhuizers naar minder populaire
plattelandsgebieden andere motieven vaak juist belangrijker zijn, zoals de
194
nabijheid van familie en vrienden en werk gerelateerde redenen. Maar op basis
van deze onderzoeken bleef het onduidelijk in hoeverre de typische ‘kwaliteit
van leven’ motieven die vaak met counterurbanisatie geassocieerd worden ook
een rol spelen voor verhuizers naar minder populair platteland. Onze studie laat
zien dat dit soort motieven ook belangrijk zijn voor verhuizers naar minder
populaire gebieden, maar dan wel vooral voor de keuze voor verhuizen naar het
platteland in het algemeen. Voor de keuze van het specifieke plattelandsgebied
zijn kenmerken van de woning het belangrijkste motief, meteen gevolgd door de
fysieke kwaliteiten van de omgeving en persoonlijke redenen, zoals intrekken bij
een partner en dichtbij familie en vrienden willen wonen. De lage huizenprijzen
in het gebied kwamen slechts als vijfde naar voren in de lijst van motieven.
Als de kenmerken en motieven van de verhuizers gecombineerd worden
komt de diversiteit binnen de groep verhuizers naar minder populaire
plattelandsgebieden nog meer naar voren. De fysieke kwaliteiten van de
omgeving trekken een groep hogeropgeleide verhuizers met hogere inkomens,
en verhuizers tussen de 35 en 64, een groep die overeenkomsten vertoont met het
stereotype beeld van counterurbanisatie. Daarnaast laat onze analyse een groep
verhuizers zien voor wie de nabijheid van familie en vrienden een belangrijk
motief is. Deze groep bestaat vaker uit terugmigranten, alleenstaanden, de
jongste en de oudste leeftijdsgroep en ook de laagste inkomensgroep. Werk
gerelateerde redenen zijn belangrijk voor hoger opgeleiden, mensen die vanuit
omliggende gemeenten verhuizen en de leeftijdsgroep 35‐44. Kenmerken van de
woning als motief verwijzen vaak naar de beschikbaarheid van een specifieke
woning. Dit motief wordt door een diverse groep verhuizers genoemd maar met
name door mensen die vanuit de stad verhuisd zijn.
Verschillende gebieden, verschillende mensen?
Om te bepalen in hoeverre deze uitkomsten echt specifiek zijn voor minder
populaire plattelandsgebieden hebben we ook een directe vergelijking gemaakt
met de kenmerken en motieven van in‐migranten naar populaire gebieden in
dezelfde regio. In hoofdstuk 2 blijkt uit migratiestatistieken dat minder populaire
gebieden een relatief groter aandeel verhuizers van binnen Noord‐Nederland
195
aantrekken, een grotere instroom vanuit plattelandsgebieden en ook een jongere
groep verhuizers. De multivariate analyse van secundaire survey data (afkomstig
uit WoON) laat zien dat mensen die naar hun huidige adres verhuisd zijn
vanwege de wens om dichtbij familie en vrienden te wonen een grotere kans
hebben om naar een minder populair gebied verhuisd te zijn. Hogeropgeleiden
en mensen die vanuit een stedelijk gebied verhuizen hebben juist een grotere
kans naar een populair gebied verhuisd te zijn. Op basis hiervan concludeerden
we dat het klassieke beeld van counterurbanisatie niet voldoet om
migratiepatronen in verschillende typen plattelandsgebieden binnen landen te
verklaren. De kenmerken van de verhuizers naar populaire gebieden lijken te
passen binnen het klassieke beeld van counterurbanisatie, terwijl minder
populaire gebieden persoonlijke redenen als een belangrijk migratiemotief delen
met perifere plattelandsgebieden in Europa.
In hoofdstuk 4 hebben we de vraag of verschillende typen
plattelandsgebieden binnen landen verschillende groepen migranten met
verschillende motieven aantrekken verder onderzocht met gebruikmaking van
nieuw verzamelde enquêtegegevens. Deze data maakten een betere analyse
mogelijk, onder meer omdat nu alleen motieven over de keuze van de
bestemming opgenomen waren, terwijl deze in de voorgaande analyse vermengd
waren met motieven om uit de vorige woonplaats te vertrekken. Ook kon nu een
variabele voor terugmigratie worden opgenomen en een meer gedetailleerde
variabele over de vorige woonplaats.
De verhuizers naar populaire en minder populaire gebieden worden
rechtstreeks vergeleken door middel van een logistische regressie analyse.
Daaruit blijkt dat de verhuizers naar populaire gebieden vaker behoren tot de
hogere inkomensgroepen en dat ze vaker hoogopgeleid zijn. Ze hebben ook
vaker eerder in de gemeente gewoond en vallen vaker in de hogere
leeftijdsgroepen. De fysieke kwaliteiten van de omgeving (bijv. natuur, ruimte),
werk gerelateerde redenen en de ligging van het gebied werden door hen vaker
genoemd als motief om naar dit gebied te verhuizen. Om de verhuizers te
karakteriseren zijn naast de meer traditionele achtergrondkenmerken ook acht
waarden die mensen meer of minder belangrijk kunnen vinden in hun leven in
196
de analyse opgenomen. Het blijkt dat in vergelijking met verhuizers naar minder
populaire gebieden deze verhuizers meer belang hechten aan de waarde ‘plezier
in het leven’.
De analyse laat zien dat verhuizers naar minder populaire gebieden vaker
vanuit elders in Noord‐Nederland verhuisd zijn dan verhuizers naar populaire
gebieden. Terwijl eerder onderzoek concludeerde dat minder populaire gebieden
vaker verhuizers over korte afstand trekken, laat onze multivariate analyse zien
dat populaire en minder populaire gebieden in dezelfde mate verhuizers vanuit
elders in Nederland aantrekken. Verhuizers naar minder populaire gebieden
vallen vaker binnen de jongste leeftijdsgroep. Minder populaire gebieden trekken
dus inderdaad een jonge groep verhuizers, ook wanneer direct vergeleken wordt
met populaire gebieden. In vergelijking met de verhuizers naar populaire
gebieden noemen ze vaker het intrekken bij een partner en de lage huizenprijzen
in het gebied als motieven om ernaartoe te verhuizen. Verhuizers naar minder
populaire gebieden noemen de fysieke kwaliteiten minder vaak, daarentegen
worden sociale kwaliteiten als vrijheid, vriendelijkheid, juist vaker door hen
genoemd als motief om naar dit gebied te verhuizen. Als we naar de waarden
kijken blijkt in vergelijking met verhuizers naar populaire gebieden, dat
verhuizers naar minder populaire gebieden meer belang hechten aan ‘rationeel
zijn’ en aan ‘afwisseling in het leven’.
Het zoekproces naar een nieuwe woning in plattelandsgebieden
In hoofdstuk 5 hebben we onderzocht hoe mensen naar een nieuwe plek om te
wonen zoeken binnen het platteland van Noord‐Nederland. Hierbij is gebruik
gemaakt van een dagboekbenadering, een nieuwe methode om het zoekproces
naar een nieuwe woning te onderzoeken. We onderscheiden drie groepen
zoekers in dit hoofdstuk. Respondenten die buiten Noord‐Nederland wonen
(verre zoekers), respondenten die al in Noord‐Nederland woonachtig zijn en die
aanvankelijk in een groter gebied in Noord‐Nederland zoeken (regionale
zoekers) en respondenten die eveneens al in Noord‐Nederland wonen en heel
lokaal zoeken (lokale zoekers). De zoekgebieden van deze groepen verschillen in
schaal en de mate waarin ze veranderen gedurende het zoekproces. Terwijl
lokale zoekers het hebben over dorpen, spreken regionale en verre zoekers in het
197
begin over gebieden. En terwijl de lokale zoekers zijn begonnen met zoeken in
hun eigen of een naburig dorp en dat blijven doen verandert het zoekgebied van
regionale en verre zoekers veel meer. Het belang van lokale bindingen voor
lokale zoekers is een belangrijke verklaring voor deze verschillende
zoekpatronen. De minder sterke voorkeur voor een specifieke plattelandslocatie
maakt het voor de regionale en verre zoekers makkelijker om hun zoekgebied
aan te passen. Dezelfde soort lokale bindingen die lokale zoekers in de buurt van
hun huidige woonplaats houden, werken ook als een beperking in het
zoekproces van verre zoekers. Financiële beperkingen stuurden het zoekgebied
van verre zoekers vaak richting het noorden met zijn lagere huizenprijzen, nadat
andere plattelandsgebieden uit hun zoekgebied waren verdwenen vanwege te
hoge prijzen. Voor lokale en regionale zoekers spelen financiële beperkingen
meer een rol in de keuze voor specifieke huizen dan in de keuze van
zoekgebieden. Tenslotte laten de resultaten zien dat het niet de grotere kennis
van de lokale huizenmarkt is die het voor lokale zoekers makkelijker maakt de
woning van hun voorkeur te vinden. Alle zoekers maken gebruik van het
internet om informatie krijgen over hun mogelijkheden op de huizenmarkt. Ze
hebben daardoor vergelijkbare kennis over wat er beschikbaar is. Wat wel
verschilt is hun kennis van de sociale kwaliteiten van een gebied. Lokale en
regionale zoekers zijn in staat onderscheid te maken tussen verschillende dorpen
en zelfs straten. Verre zoekers ontwikkelden verschillende strategieën om deze
achterstand in te halen, zoals rondrijden door het gebied en contact zoeken met
de buren bij het bekijken van een woning. Het belang dat aan deze sociale
kwaliteiten wordt gehecht en ook het belang van het gevoel bij een gebied of een
woning zorgt ervoor dat het internet ‘oude’ zoekmethoden als rondrijden en
huizen bezichtigen niet heeft vervangen. Het lijkt er wel op dat huizenwebsites
zoals Funda in Nederland het makkelijker maken voor verre zoekers om nieuwe
potentiële zoekgebieden te verkennen, wat soms resulteert in een aanpassing van
het zoekgebied.
Representaties van gebieden spelen op verschillende manieren een
belangrijke rol in het zoekproces. Direct en indirect contact met een gebied blijkt
essentieel te zijn om het gebied op te nemen in het zoekgebied. Maar het is ook zo
198
dat kennis van een gebied niet noodzakelijkerwijs leidt tot een positief beeld van
een gebied. Zowel lokale als regionale zoekers noemden gebieden binnen de
provincie of specifieke dorpen waar ze niet willen wonen, en verre zoekers
verwezen naar hele provincies die ze links lieten liggen op basis van hun beeld
ervan. Wanneer ze zoekactiviteiten ontplooien ontdekken zoekers binnen het
grotere gebied vaak gebieden die verschillende beelden en gevoelens oproepen.
Wat opvalt is dat de respondenten vaak verwezen naar sociale aspecten, zoals het
type mensen dat ergens woonde, wanneer ze het hadden over kenmerken op
basis waarvan plekken werden uitgesloten van het zoekproces.
Een nieuwe blik op counterurbanisatie
Onze resultaten bevestigen het idee dat counterurbanisatie ‘een complex en
gedifferentieerd fenomeen’ is, zowel wanneer je kijkt naar migratie naar minder
populaire gebieden als wanneer gekeken wordt naar migratiestromen naar
verschillende typen plattelandsgebieden. Zoals eerder gezegd liet de analyse
waarin de motieven en kenmerken van verhuizers aan elkaar gelinkt werden de
diversiteit zien binnen de verhuizers naar minder populaire gebieden. Op een
vergelijkbare manie zou ook in populaire gebieden de aanwezigheid van
verschillende groepen verhuizers geanalyseerd kunnen worden.
Het vergelijken van migratiestromen naar verschillende typen
plattelandsgebieden binnen één land laat zien dat counterurbanisatie inderdaad
meer is dan het klassieke beeld suggereert. Een bredere opvatting van wat
counterurbanisatie inhoudt is dus noodzakelijk. Dat idee is al eerder naar voren
gebracht, maar vaak wordt dan het traditionele beeld geassocieerd met populaire
plattelandsgebieden, terwijl het beeld van andere groepen die naar het platteland
trekken met meer praktische motieven vaak gekoppeld wordt aan minder
populaire gebieden. Onze uitkomsten laten zien dat het onderscheid niet zo
scherp te maken is. In onze analyse blijkt bijvoorbeeld wel dat populaire
gebieden meer verhuizers met hogere inkomens trekken, maar beide typen
gebieden trekken verhuizers uit de stad. Ook voor verhuizers naar populaire
gebieden spelen meer instrumentele redenen gerelateerd aan werk en locatie een
rol. En terwijl de fysieke kwaliteiten belangrijk zijn voor verhuizers naar
populaire gebieden noemen verhuizers naar minder populaire gebieden vaker de
199
sociale kwaliteiten. Specifieke ‘plattelandskwaliteiten’ doen er dus zeker ook toe
voor verhuizers naar minder populaire gebieden. Toekomstig onderzoek naar
migratie naar het platteland doet er dus goed aan voorzichtig te zijn met het
gebruiken van een te simpele opvatting van counterurbanisatie, zowel in het
algemeen als met betrekking tot verschillende typen platteland binnen landen.
Meer aandacht nodig voor de sociale kwaliteiten van plattelandsgebieden
In de context van het platteland ligt de nadruk vaak op de invloed van landschap
en ligging op woonplaatskeuze. Onze uitkomsten suggereren dat meer aandacht
nodig is voor de rol van sociale kenmerken van plattelandsgebieden in de keuze
voor een nieuwe plek om te wonen. In hoofdstuk 5 bleek dat respondenten vaak
sociale aspecten noemden wanneer ze spraken over kenmerken op basis waarvan
ze gebieden uitsloten van hun zoekproces. In een stedelijke context wordt de rol
van percepties van sociale kenmerken van gebieden in het proces van het zoeken
naar een nieuwe woning wel erkend. Zo is bijvoorbeeld bekend dat de reputatie
van een stadsbuurt het meest beïnvloed wordt door de sociaal‐economische en
etnische samenstelling van de buurt, terwijl fysieke en functionele kenmerken
veel minder van belang zijn.
In hoofdstuk vier vonden we dat de sociale kwaliteiten van de omgeving,
zoals vrijheid en vriendelijkheid, voor verhuizers naar minder populaire
gebieden een belangrijker motief waren om naar dit specifieke plattelandsgebied
te verhuizen dan voor verhuizers naar populaire gebieden. Voor verhuizers naar
populaire gebieden waren de fysieke kwaliteiten van de omgeving juist
belangrijker in de keuze voor hun huidige woongebied dan voor verhuizers naar
minder populaire gebieden. Hieruit blijkt dat het nuttig is een onderscheid te
maken tussen fysieke en sociale aspecten van de plattelandsomgeving om
verhuizen naar verschillende typen platteland te verklaren. Het roept de vraag
op of minder populaire gebieden wellicht meer sociale kwaliteiten te bieden
hebben. Helaas zijn sociale kwaliteiten lastiger te operationaliseren dan de
fysieke kwaliteiten die vaak gebruikt worden om de aantrekkelijkheid van
plattelandsgebieden vast te stellen. Om deze vraag te kunnen beantwoorden is
dan ook nader onderzoek nodig om te bekijken welke variabelen relevante
indicatoren kunnen zijn voor de sociale kwaliteiten van plattelandsgebieden.
200
Beleidsoverwegingen
Op basis van de resultaten van dit onderzoek kan een aantal suggesties voor
beleidsmakers in minder populaire plattelandsgebieden geformuleerd worden.
De uitkomsten laten zien dat het aandeel lokale en regionale verhuizers groot is.
Dit maakt het de moeite waard te investeren in het behouden van deze groepen
voor het gebied, in plaats van veel moeite te doen om mensen van verder weg
aan te trekken. De interviews met verre zoekers lieten ook zien hoe sociale en
werk gerelateerde bindingen aan hun huidige woonomgeving het soms lastig
maken hun verhuisplannen te realiseren. Verder bleek dat de kwaliteit van de
omgeving zeker ook een belangrijke rol speelt bij verhuizen naar minder
populaire plattelandsgebieden. Dit betekent dat het belangrijk is kwaliteiten als
rust, ruimte, natuur en landschap ook in minder populaire gebieden te bewaren,
om aantrekkelijk te blijven voor in‐migranten. Hoofdstuk 5 laat zien dat
representaties van gebieden op verschillende manieren een belangrijke rol spelen
in het zoekproces. Direct of indirect contact blijkt essentieel te zijn om het gebied
op te nemen in het zoekgebied. Daarmee kan het bevorderen van toerisme in een
gebied een indirecte manier zijn om nieuwe inwoners aan te trekken, en ook
experimenten als het uitdelen van gratis treinkaartjes en ‘proefwonen’ kunnen
van waarde zijn voor gebieden die mensen willen aantrekken. Maar kennis van
een gebied leidt niet altijd tot een positief beeld. Lokale en regionale zoekers
noemden gebieden binnen de provincie of specifieke dorpen waar ze niet zouden
willen wonen en verre zoekers hele provincies, op basis van hun perceptie van
het gebied of dorp. Negatieve percepties zijn lastig positief te beïnvloeden met
beleidsmaatregelen. Niettemin is het belangrijk ze ook niet in negatieve zin te
beïnvloeden. De discussie over bevolkingskrimp in Nederland leidt tot beelden
in de media van dichtgespijkerde ramen en lege winkels. Soms lijkt het alsof
beleidsmakers deze beelden willen versterken, bijvoorbeeld om hun verzoek om
financiële steun bij de landelijke overheid te ondersteunen. Voor het aantrekken
van nieuwe inwoners is dit geen goede strategie. Het is belangrijk om te
erkennen dat net als in stedelijke gebieden reputaties van gebieden een rol spelen
in de woonplaatskeuze en dat deze reputaties hardnekkig zijn. Daarom is
revitaliseren van een gebied niet voldoende, ook het beeld naar buiten toe moet
veranderen.
201
Tenslotte laten de resultaten van hoofdstuk 4 zien dat de waarden die
mensen belangrijk vinden in het leven verschillen voor migranten naar
verschillende typen plattelandsgebieden, en dat deze waarden lijken samen te
hangen met de meer abstracte ideeën of representaties die mensen hebben bij
deze gebieden. Nader onderzoek hiernaar is noodzakelijk, maar het gebruiken
van dit soort waarden in de marketing van plattelandsgebieden zou nieuwe
mogelijkheden kunnen bieden, net zoals dat gebeurt in stedelijke gebieden. De
analyse liet zien dat minder populaire gebieden staan voor een rationele keuze of
zelfs een avontuurlijke keuze: minder populaire gebieden als plekken om een
leven vol uitdaging, vernieuwing en verandering te leiden. In het algemeen is het
belangrijk voor beleidsmakers in plattelandsgebieden om zich te realiseren dat
migratie naar het platteland niet alleen een zaak is van welvarende stedelingen
op zoek naar een plattelandsidylle. De uitkomsten van dit onderzoek laten zien
dat ‘de’ verhuizer naar het platteland niet bestaat. Het is essentieel om te
erkennen dat er verschillende groepen verhuizers zijn met verschillende
motieven en dat deze verschillende groepen op verschillende manieren benaderd
moeten worden.